DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
This office action is in response to the amendment filed on 06/29/2026. Claims 1-2 are pending. Claim 1 is independent.
Claim Objections
Applicant's amendment to claims corrects some of previous objections; therefore, some of previous objections are withdrawn. The remaining objections are shown below.
Claim 1-2 are objected to because of the following informalities:
In Claim 1, lines 18-23, "… evaluate the data related to the at least one machining condition … based on a result of the simulation … update the machining information database … based on a result of an evaluation …" appears to be , "… evaluate the data related to the at least one machining condition … based on a result of the simulation … update the machining information database … based on a result of the evaluation …" to indicate that "updating" is based on a result of "evaluating the data related to the at least one machining condition" recited earlier and not a result of any evaluation.
Appropriate correction is required.
Claim Interpretation
Applicant's amendment to claims avoids the limitations being interpreted under 35 U.S.C. 112(f).
Claim Rejections - 35 USC § 112
Applicant's amendment to claims corrects some of previous rejections; therefore, some of previous rejections are withdrawn. Applicant's amendment to claims also raises new issues; therefore, the remaining rejections are shown below.
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-2 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "… for estimating a machining condition in a wire electric discharge machine … stores machining information data in which data related to a machining condition satisfying a machining content and a required specification … derive … data related to at least one machining condition estimated to satisfy a desired machining content and the required specification …" in lines 1-, which rendering the claim indefinite because it is unclear (1) whether the first two instances of ".
Claim 2 is rejected for fully incorporating the deficiency of their respective base claims.
Claim 2 recites the limitation "… learn … a correlation among the data related to the machining content, the data related to the required specification, and the data related to the machining condition … derive, based on a learning result of the correlation, data related to the at least one machining condition estimated to satisfy the desired machining content and the required specification" in lines , which rendering the claim indefinite because ".
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2 are rejected under 35 U.S.C. 103 as being unpatentable over TSUNODA et al. (US 2020/0279158 A1, pub. date: 09/03/2020), hereinafter TSUNODA in view of GOYA et al. (US 2020/0293021 A1, pub. date: 09/17/2020)., hereinafter GOYA, NAKASHIMA (US 2018/0281091 A1, pub. date: 10/04/2018), hereinafter NAKASHIMA, and Akemura (US Patent 5,742,018, Date of Patent: 04/21/1998), hereinafter Akemura.
Independent Claim 1
TSUNODA discloses a machining condition estimation device for estimating a machining condition in a (TSUNODA, ¶¶ [0068], [0070], [0072]-[0073], [0075]-[0077], [0089], [0091] ,[0093], and [0096]-[0101] with FIGS. 5 and 7: estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type, when the state of machining by the machine tool 2 is given; an estimation unit 120 is capable of carrying out estimation processing that is based on the state data S acquired from the machine tool 2 and that is demanded for determination of at least either of the machining conditions and the machining parameters which are more appropriate and which correspond to the machining type in the acquired state; the machining condition adjustment device 1 includes a configuration for estimation that is demanded when the machine learning device 100 estimates at least either of machining conditions and machining parameters for machining by each machine tool; the estimation unit 120 carries out the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model selected by the learning model selection unit 105, based on the state data S produced by the preprocessing unit 36; estimating at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and controlling the machining operation for a workpiece by the machine tool 2 based on at least either of the machining conditions and the machining parameters that have been estimated; estimating at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and controlling the machining operation for a workpiece by the machine tool 2 based on at least either of the machining conditions and the machining parameters that have been estimated), the machining condition estimation device comprising:
a processor (TSUNODA, ¶ [0025] and [0032] with 11 and 101 in FIG.1: a CPU (Central Processing Unit) 11 included by the machining condition adjustment device 1 is a processor that generally controls the machining condition adjustment device 1; the machine learning device 100 includes a processor 101 which controls the whole machine learning device 100); and
a machining information database that stores machining information data in which data related to a machining condition satisfying a machining content and a required specification are associated with data related to the machining content and data related to the required specification (TSUNODA, ¶¶ [0026] and [0035] with FIGS. 1-2: various types of data (information on tools such as types of the tools, information on cutting conditions such as spindle speed, feed speed, and cutting depth, information on workpieces such as materials and shapes of the workpieces, power to be consumed by each motor, dimension values and surface quality of portions of machined workpieces and temperatures of portions of the machine tool that have been measured by sensors 3, or the like) that have been acquired from units of the machining condition adjustment device 1, the machine tool, the sensors 3, and the like have been stored; examples of the data that are acquired by the control unit 32 from the machine tool 2 and the sensors 3 and that are outputted to the data acquisition unit 34 include the information on the tools such as the types of the tools, the information on the cutting conditions such as the spindle speed, the feed speed, and the cutting depth, the information on machining parameters, the information on the workpieces such as the materials and the shapes of the workpieces, the information such as the power to be consumed by each motor and the temperatures of the portions of the machine tool, the information on results of the machining such as the dimensions and the profile irregularity of the portions of the machined workpieces, and the like; ¶¶ [0038]-[0039] with 52 and 56 in FIG. 2: the data acquisition unit 34 stores, in the acquired data storage unit 52, the data, related to the state and the result of the machining by the machine tool 2, inputted from the control unit 32, the data related to the result of the machining, inputted by an operator from the display/MDI unit 70, the machining type determined by the machining type determination unit 33, and the like; the data acquisition unit 34 associates the data, related to the state and the result of the machining by the machine tool 2, inputted from the control unit 32, the data related to the result of the machining, inputted by the operator, the machining type determined by the machining type determination unit 33, and the like with one another and stores the associated data as acquired data in the acquired data storage unit 52; the priority condition setting unit 37 receives a priority condition for each machining type in machining of a workpiece and stores the priority condition in the priority condition storage unit 56; the priority condition setting unit 37 displays a UI screen for the setting of the priority condition for each machining type on the display/MDI unit 70, for instance, acquires the priority condition for each machining type that is set up by operation by the operator through the UI screen, produces the priority condition data that is data in which the machining type is associated with the priority condition, and stores the priority condition data in the priority condition storage unit 56; ¶¶ [0040]-[0041] and [0047] with FIG. 3: the machining types may be defined as machining processes of carrying out machining with given purposes, such as roughing, finishing, and profile machining; the priority conditions may be definable with use of demanded particulars for the machining such as "high cycle time", "energy saving", "high quality machining", "machining accuracy", "extension of lives of components of machine tool", "extension of tool lives", "extension of lives of peripheral devices", "reduction in maximum peak power for machine tool", "increase in yield rate of machined products", "optimization of shape and size of chips", "reduction of vibrations, noise, electromagnetic noise, and heat generated from machine tool and peripherals", or "reduction in heat generation in machine tool" (i.e., specification); a conditional expression that defines a specific machining condition or a specific condition for a machining parameter, measured data, the data related to the result of machining, or the like may be defined in advance for each of the demanded particulars and may be stored in the priority condition storage unit 56 so that the conditional expression may be referred to when a determination concerning each of the demanded particulars is made; a conditional expression, such as "pitch error < Errpit", for a machining condition, a machining parameter, measured data, the data related to the result of machining, or the like may be directly defined as a priority condition (i.e., specification); the priority condition data may be data in which a plurality of priority conditions are associated with one machining type; alternatively, the priority condition data may be data in which priorities are assigned to a plurality of priority conditions; ¶¶ [0042]-[0046], [0063]-[0066], and [0084]-[0087] with FIGS. 2 and 4: the preprocessing unit 36 produces learning data to be used for machine learning based on the data acquired by the data acquisition unit 34 and the priority condition data stored in the priority condition storage unit 56; for reinforcement learning, the preprocessing unit 36 produces a set of state data S and determination data D in given formats in the learning, as the learning data; the state data S that is produced by the preprocessing unit 36 includes at least tool data S1 including information on tools to be used for machining of workpieces by the machine tool 2 and either of machining condition data S2 including information on the machining conditions in the machining of the workpieces by the machine tool 2 and machining parameter data S3 including parameter information related to the machining of the workpieces by the machine tool 2 and machining parameter data S3 including parameter information related to the machining of the workpieces by the machine tool 2; the tool data S1 (i.e., machining content) is defined as data strings indicating types and materials of the tools to be used for the machining of the workpieces by the machine tool 2; the types of the tools may be classified into cutting tool, milling cutter, drilling tool, or the like in accordance with shapes of the tools or usage in the machining and may be expressed as numerical values that each have a unique identification; the materials of the tools, such as high speed steel and cemented carbide, may be expressed as numerical values that each have a unique identification; the tool data S1 may be produced through acquisition of the information on the tools set for the machining condition adjustment device 1 and the machine tool 2 by an operator and on the basis of the acquired information on the tools; the machining condition data S2 is defined as data strings including the machining conditions such as the spindle speed, the feed speed, and the cutting depth based on the settings or the instructions for the machining of the workpieces by the machine tool 2, as elements; the machining parameter data S3 is defined as data strings including control parameters for the machine that are referred to for the machining of the workpieces by the machine tool 2, as elements; the control parameters are parameters such as control time constants of motors for control over the machine tool 2, parameters related to control over the peripherals or the like, and so forth; ¶¶ [0057], [0071], [0078], [0092], and [0102] with FIGS. 4-7: the preprocessing unit 36 may further produce workpiece data S4 indicating information on workpieces to be machined by the machine tool 2, as the state data, in addition to the tool data S1, the machining condition data S2, and the machining parameter data S3; the workpiece data S4 (i.e., machining content) is defined as data strings indicating materials of the workpieces to be machined by the machine tool 2; the materials of the workpieces, such as aluminum and iron, may be expressed as numerical values that each have a unique identification (i.e., machining content); ¶¶ [0062], [0067]-[0070], [0075]-[0077], [0083], [0088]-[0091], and [0096]-[0101] with FIGS. 4-7: produce the set of the state data S and the label data L, as the learning data, from only the acquired data satisfying a priority condition associated with the machining type in a state in which each acquired data is acquired, among the data acquired by the data acquisition unit 34 (and stored in the acquired data storage unit 52); it is only the learning data produced based on the data acquired with satisfaction of the priority condition for each machining type; learn at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type, in association with the state of machining by the machine tool 2; estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type, when the state of machining by the machine tool 2 is given), wherein the processor is configured to:
derive, based on the machining information data stored in the machining information database, data related to at least one machining condition estimated to satisfy a desired machining content and the required specification (TSUNODA, ¶¶ [0006]-[0007]: adjust at least either of machining conditions and machining parameters in consideration of particulars demanded in accordance with a machining type of a workpiece in a machine tool; ¶ [0032] wit FIG. 1: the machine learning device 100 is capable of observing information (the information on the tools such as the types of the tools, the information on the cutting conditions such as the spindle speed, the feed speed, and the cutting depth, the information; on the workpieces such as the materials and the shapes of the workpieces, the power to be consumed by each motor, the dimension values and the surface quality of the portions of the machined workpieces and the temperatures of the portions of the machine tool that have been measured by the sensors 3, or the like, for instance) that may be acquired in the machining condition adjustment device 1, through the interface 21; ¶¶ [0048] and [0053] with FIG.2: the learning model selection unit 105 selects a learning model corresponding to the machining type inputted from the preprocessing unit 36 and causes the selected learning model to be used for the learning by the learning unit 110 and decision making by the decision making unit 122; the learning model may be configured so as to carry out more effective learning and inferencing by using a so-called deep learning technique with use of a neural network that forms three or more layers; the learning model is used for determination of the adjustment behavior for at least either of the machining conditions and the machining parameters by the decision making unit 122; ¶¶ [0068], [0070], [0072], and [0075]-[0077] with FIGS. 4-5: the learning model generated by the learning unit 110 may be used for estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type, when the state of machining by the machine tool 2 is given; generates a plurality of learning models in which at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and the state of machining by the machine tool 2 have been learned in association; with use of the plurality of learning models generated in this manner, an estimation unit 120 is capable of carrying out estimation processing that is based on the state data S acquired from the machine tool 2 and that is demanded for determination of at least either of the machining conditions and the machining parameters which are more appropriate and which correspond to the machining type in the acquired state; the machining condition adjustment device 1 includes a configuration for estimation that is demanded when the machine learning device 100 estimates at least either of machining conditions and machining parameters for machining by each machine tool; in a stage of the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model by the machine learning device 100, the preprocessing unit 36 carries out the conversion (such as digitization or sampling) into the unified format that is handled in the machine learning device 100, based on the data acquired by the data acquisition unit 34 and the priority condition data stored in the priority condition storage unit 56; the state data S in a given format that is used for the estimation by the machine learning device 100 is produced from the converted data and the produced state data S, together with the machining type, is outputted to the machine learning device 100; the estimation unit 120 carries out the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model selected by the learning model selection unit 105, based on the state data S produced by the preprocessing unit 36; in the estimation unit 120, the state data S inputted from the preprocessing unit 36 is inputted into the learning model generated (having the parameters determined) by the learning unit 110 and at least either of the machining conditions and the machining parameters satisfying the priority condition associated with the machining type are thereby estimated and outputted; the machining condition adjustment device 1 is capable of estimating at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and controlling the machining operation for a workpiece by the machine tool 2 based on at least either of the machining conditions and the machining parameters that have been estimated; ¶¶ [0089], [0091], and [0093]-[0097] with FIGS. 6-7: the learning model generated by the learning unit 110 may be used for estimation of at least either of the machining conditions and the machining parameters satisfying the priorities corresponding to the machining type in the machining by the machine tool 2; generates a plurality of learning models in which the distribution of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type has been learned. With use of the plurality of learning models generated in this manner, the estimation unit 120 is capable of carrying out estimation processing that is based on the state data S acquired from the machine tool 2 and that is demanded for determination of at least either of the machining conditions and the machining parameters which are more appropriate and which correspond to the machining type in the acquired state; the machining condition adjustment device 1 includes a configuration that is demanded when the machine learning device 100 estimates at least either of machining conditions and machining parameters for machining by each machine tool; in a stage of the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model by the machine learning device 100, the preprocessing unit 36 carries out the conversion (such as digitization or sampling) into the unified format that is handled in the machine learning device 100, based on the data acquired by the data acquisition unit 34 and the priority condition data stored in the priority condition storage unit 56; the state data S in a given format that is used for the estimation by the machine learning device 100 is produced from the converted data and the produced state data S, together with the machining type, is outputted to the machine learning device 100; the estimation unit 120 carries out the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model selected by the learning model selection unit 105, based on the state data S produced by the preprocessing unit 36; in the estimation unit 120 of the embodiment, at least either of the machining conditions and the machining parameters satisfying the priority condition are estimated and outputted based on a position of the state data S inputted from the preprocessing unit 36 in a distribution of data included in the learning model generated by the learning unit 110); execute a machining of the on the data related to the at least one machining condition derived by the processor; evaluate the data related to the at least one machining condition derived by the processor based on a result of the machining (TSUNODA, ¶¶ [0026]-[0027] and [0032] with FIG.1: dimension values and surface quality of portions of machined workpieces and temperatures of portions of the machine tool that have been measured by sensors 3; the sensors 3 are used for measurement of errors with respect to design data, profile irregularity, or the like in portions of a workpiece machined by the machine tool; ¶¶ [0035]m [0040], and [0047] with FIGS. 1-2: the control unit 32 controls machining operation by the machine tool 2 and measurement operation by the sensors 3, based on a control program 54; the control unit 32 has functions for general control, demanded for control over the portions of the machine tool 2, such as output of travel instructions at every control cycle to the servo motors 50 (FIG. 1) that move the axes included in the machine tool 2 and to the spindle motor 62 (FIG. 1) based on the control program 54; in addition, the control unit 32 outputs instructions to carry out the measurement operation to the sensors 3; the control unit 32 has functions for general control, demanded for control over the portions of the machine tool 2, such as output of travel instructions at every control cycle to the servo motors 50 (FIG. 1) that move the axes included in the machine tool 2 and to the spindle motor 62 (FIG. 1) based on the control program 54; in addition, the control unit 32 outputs instructions to carry out the measurement operation to the sensors 3; the control unit 32 receives data related to a state and a result of machining by the machine tool 2, from the machine tool 2 and the sensors 3, and outputs the data to the data acquisition unit 34; the state data S is produced based on the data acquired from the machine tool 2 (and the sensors 3) during the drilling/ profile machining; produce the determination data D corresponding to the state data S, based on data for determination of the surface quality or the like measured by the sensors 3; ¶¶ [0042] and [0049]-[0052] with FIG. 2: on condition that the machine learning device 100 carries out the reinforcement learning, e.g., the preprocessing unit 36 produces a set of state data S and determination data D in given formats in the learning, as the learning data; the reinforcement learning is the technique in which a cycle including observing a current state (that is, input) of an environment where a learning object exists, executing given behavior (that is, output) in the current state, and conferring some reward for the behavior is iterated by a trial-and-error method and in which a measure (the adjustment behavior for at least either of the machining conditions and the machining parameters in the machine learning device 100 of the application) that maximizes total of such rewards is learned as an optimal solution; in the Q-learning by the learning unit 110, the reward R may be determined based on the priority condition associated with the machining type stored in the priority condition storage unit 56; the state data S which is the current learning object is based on the data acquired during the drilling; the reward R may be positive (plus) when the pitch error as the determination data is smaller than the threshold Errpit stored in the priority condition storage unit 56 or may be negative (minus) when the pitch error is equal to or greater than the threshold Errpit; the pitch error magnitude of the positive or negative reward may be changed in accordance with a deviation of the pitch error from the threshold Errpit; the reward R may be calculated with use of an expression in which the plurality of priority conditions are combined; the state data S which is the current learning object is based on the data acquired during the finishing, e.g., a given formula for reward calculation in which data related to the shape precision and data related to the cycle time are used may be defined in advance and the reward R may be calculated with use of the formula for reward calculation; the learning unit 110 may use a neural network as a value function Q (learning model) and may be configured so as to use the state data S and behavior a as input of the neural network and so as to output a value (result y) of the behavior a in a pertinent state; ¶¶ [0055]-[0056]: the decision making unit 122 determines the optimal solution of the adjustment behavior for at least either of the machining conditions and the machining parameters with use of the learning model selected by the learning model selection unit 105 based on the state data S inputted from the preprocessing unit 36 and outputs the determined adjustment behavior for at least either of the machining conditions and the machining parameters; the decision making unit 122 inputs the state data S (the tool data S1, the machining condition data S2, and the machining parameter data S3) inputted from the preprocessing unit 36 and the adjustment behavior for at least either of the machining conditions and the machining parameters (a combination of adjustment for the feed speed, adjustment for the spindle speed, and the like, or change in setting of the parameters) as input data into the learned model updated (having the parameters determined) through the reinforcement learning by the learning unit 110, so that the reward in case where the pertinent behavior is executed in the current state is calculated; the calculation of the reward in the decision making unit 122 is carried out for the adjustment behavior for at least either of the machining conditions and the machining parameters that may be currently adopted; through a comparison among a plurality of calculated rewards, the adjustment behavior for at least either of the machining conditions and the machining parameters that results in calculation of the largest reward is determined as the optimal solution; the optimal solution of the adjustment behavior for at least either of the machining conditions and the machining parameters determined by decision making unit 122 is inputted into the control unit 32 and is used for determination of at least either of the machining conditions and the machining parameters in actual machining; the machining condition adjustment device 1 can adjust appropriately for at least either of the machining conditions and the machining parameters in accordance with particulars demanded by an operator during the machining of a workpiece by the machine tool 2; ¶¶ [0098]-[0101] with FIG. 7: the estimation unit 120 calculates a distance between each data set (cluster) in the distribution of the data included in the learning model generated by the learning unit 110 and the position of the state data S inputted from the preprocessing unit 36 and estimates that the current machining conditions or machining parameters satisfy the priority condition in the current machining type, in case where the distance between the position of the state data S inputted from the preprocessing unit 36 and a nearest data set C1n is equal to or shorter than a predetermined and given threshold Distth1, for instance; the estimation unit 120 estimates that the priority condition in the current machining type is not satisfied because of at least either of the current machining conditions and machining parameters, in case where the distance between the position of the state data S inputted from the preprocessing unit 36 and the nearest data set C1n is longer than the predetermined and given threshold Distth1, for instance; the estimation unit 120 adjusts one machining condition or machining parameter or a plurality of machining conditions or machining parameters within the state data S inputted from the preprocessing unit 36 in accordance with a predetermined and given rule so that the distance to the data set C1n may be made equal to or shorter than the predetermined and given threshold Distth1; the predetermined rule for such adjustment may be a rule by which a given machining condition or a given machining parameter is fixedly adjusted, for instance; the rule may provide that the adjustment shall be made so that the distance to the data set C1n may be made equal to or shorter than the predetermined and given threshold Distth1 by a smallest adjustment amount; there may be a rule that a given machining condition or a given machining parameter shall be excluded from objects of the adjustment; thus the estimation unit 120 estimates and outputs at least either of the machining conditions and the machining parameters satisfying the priority condition, based on the current state data S and the learning model; the estimation unit 120 may output an instruction to stop the machining, in case where the distance between the position of the state data S inputted from the preprocessing unit 36 and the nearest data set C1n is longer than a predetermined and given threshold Distth2 (threshold Distth2 > threshold Distth1); an instruction for emergency stop may be outputted in case where the distance is longer than a predetermined and given threshold Distth3 (threshold Distth3 > threshold Distth2); such a technique makes it possible to estimate that the state of machining is abnormal, in case where operation vastly different from normal operation is carried out, and to call attention of the operator; the machining condition adjustment device 1 is capable of estimating at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and controlling the machining operation for a workpiece by the machine tool 2 based on at least either of the machining conditions and the machining parameters that have been estimated); and
update the machining information database (TSUNODA, ¶¶ [0035]-[0038], [0041], and [0045]-[0047] with FIG. 2: the control unit 32 receives data related to a state and a result of machining by the machine tool 2, from the machine tool 2 and the sensors 3, and outputs the data to the data acquisition unit 34; when at least either of machining conditions and the machining parameters are outputted from the machine learning device 100, the control unit 32 controls the machining operation in the machine tool 2 with use of the machining conditions or the machining parameters outputted from the machine learning device 100, in place of machining conditions or the machining parameters based on the instructions from the control program 54 or the like; the data acquisition unit 34 associates the data, related to the state and the result of the machining by the machine tool 2, inputted from the control unit 32, the data related to the result of the machining, inputted by the operator, the machining type determined by the machining type determination unit 33, and the like with one another and stores the associated data as acquired data in the acquired data storage unit 52; the priority condition setting unit 37 may be made capable of defining the priority condition for each machining type in the control program 54 and setting the priority condition for each machining type that is read from the control program 54; the values of the respective machining conditions are set by instructions from the control program 54 or as default values for control and thus may be produced through acquisition of the instructions or the default values; ¶ [0055] with FIG. 2: the optimal solution of the adjustment behavior for at least either of the machining conditions and the machining parameters determined by decision making unit 122 is inputted into the control unit 32 and is used for determination of at least either of the machining conditions and the machining parameters in actual machining; ¶¶ [0062]-[0066] with FIG. 4 and ¶¶ [0086]-[0087] with FIG. 6: produce the set of the state data S and the label data L, as the learning data, from only the acquired data satisfying a priority condition associated with the machining type in a state in which each acquired data is acquired, among the data acquired by the data acquisition unit 34 (and stored in the acquired data storage unit 52); machining condition label data L1 labeled with information on the machining conditions in the machining of a workpiece by the machine tool 2 in the state of machining in which the state data S is acquired or machining parameter label data L2 including parameter information related to the machining of the workpiece by the machine tool 2; the machining condition label data L1 is defined as data strings including the machining conditions such as the spindle speed, the feed speed, and the cutting depth based on the settings or the instructions for the machining of the workpieces by the machine tool 2, as elements; as the spindle speed, the feed speed, the cutting depth, and the like, numerical values in which values of the respective machining conditions are expressed with use of given units may be used; the values of the respective machining conditions are set by instructions from the control program 54 or as default values for control and thus may be produced through acquisition of the instructions or the default values; the machining parameter label data L2 is defined as data strings including control parameters for the machine that are referred to for the machining of the workpieces by the machine tool 2, as elements; the control parameters are parameters such as control time constants of motors for control over the machine tool 2, parameters related to control over the peripherals or the like, and so forth; as the machining parameter label data L2, the parameters set during the machining may be acquired; ¶ [0076] with FIG. 5 and ¶ [0097] with FIG. 7: a result estimated by the estimation unit 120 is outputted to the control unit 32)(TSUNODA, ¶¶ [0049], [0053], and [0055] with FIG.2: the learning unit 110 updates the learning model selected by the learning model selection unit 105 so as to learn adjustment behavior for at least either of the machining conditions and the machining parameters with respect to the state and the result of the machining by the machine tool 2, in accordance with a publicly known reinforcement learning technique and stores the updated learning model in the learning model storage unit 130; the learning model updated by the learning unit 110 is stored in the learning model storage unit 130 provided on the nonvolatile memory 104; the learned model updated (having the parameters determined) through the reinforcement learning by the learning unit 110, so that the reward in case where the pertinent behavior is executed in the current state is calculated; ¶¶ [0067]-[0068] with FIG. 4: the learning unit 110 updates the learning model selected by the learning model selection unit 105 so as to learn at least either of the machining conditions and the machining parameters satisfying the priority condition with respect to the state of machining by the machine tool 2, in accordance with a publicly known supervised learning technique and stores the updated learning model in the learning model storage unit 130; the learning unit 110 updates the learning model so as to learn at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type, in association with the state of machining by the machine tool 2; ¶¶ [0088]-[0089] with FIG. 6: the learning unit 110 updates the learning model selected by the learning model selection unit 105 so as to learn the distribution of at least either of the machining conditions and the machining parameters satisfying the priority condition in the machining by the machine tool 2, in accordance with a publicly known unsupervised learning technique and stores the updated learning model in the learning model storage unit 130; the learning unit 110 updates the learning model so as to learn the distribution of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type in the machining by the machine tool 2).
TSUNODA fails to explicitly disclose wherein (1) the electric machine includes a wire electric discharge machine; (2) a simulation unit that executes a simulation of a wire electric discharge machine based on the machining condition derived by the machining condition derivation unit; (3) a machining condition evaluation unit that evaluates the machining condition based on a result of the simulation by the simulation unit; and (4) update the machining information database by adding, to the machining information database, machining information data created.
GOYA teaches a system and a method for determining machining condition (GOYA, ¶¶ [0001]-[0003]), wherein a simulation unit that executes a simulation of a electric machine based on the machining condition derived by the machining condition derivation unit (GOYA, ¶¶ [0029]-[0032] with FIGS. 1 and 2A-2B: a simulation system 1 provides a simulation function of simulating machining by machine tools 3, 3a, and 3b and calculating a machining result assumed when the machine tool 3 or the like performs machining; with respect to machining performed by the machine tool 3, the simulation device 10 simulates machining by the machine tool 3, and calculates a machining result by inputting a machining detail and a setting condition to a simulation model for machining; then, the simulation device 10 provides the machining result to a user, wherein the machining detail is a request and a specification of machining for an object to be machined, and the setting condition is an operating condition (the machining condition) of the machine tool 3 set on the machine tool 3 for performing appropriate machining; an example of the machining detail in FIG. 2A includes the machining detail showing that a tapered hole in which hole diameter of an inlet is "50 μm" and hole diameter of an outlet is "60 μm" are formed on a member which is made of "Si" and has a panel thickness of "400 μm"; the machining detail includes not only items related to a shape such as the hole diameter and a hole depth but also items related to quality; the items related to the quality include, e.g., a cross-sectional area of deteriorated layers, a height of burrs, a size of deposits, and surface roughness; FIG. 2B shows an example of the setting condition in a case where the machine tool 3 is a laser machining apparatus; the setting condition of the laser machining apparatus include, e.g., power of a laser to be output, piercing time, rotation speed of a revolving head of the laser, XY shaft feeding speed, defocus amount, a taper angle, gas pressure of an assist gas, a gas type, a revolving diameter of the laser, and the like; as shown in the figs, a value of each item of the setting condition is given in a range; the range of each item is a range determined in consideration of influence according to disturbance such as installation environment of the machine tool and an individual difference (the material) of the object to be machined; ¶¶ [0035]-[0038] with FIG. 1: the input/output unit 11 acquires, for actual machining performed in the machine tool 3, machining detail information that is information indicating the machining detail, setting condition information that is information indicating the setting condition in machining, and machining result information that is information indicating the machining result; the machining result information includes, e.g., information on an image of photographing the object to be machined after machining and the shape or the quality obtained by analyzing the image, and information on a measurement result of a prescribed portion of the object to be machined after machining; the simulation execution unit 12 inputs the machining detail information and the setting condition information, and calculates the machining result by a prescribed simulation model; the machining result calculated by the simulation execution unit 12 is referred to as simulation result information; the simulation result information includes information on the shape and the quality of a machining product, such as a two-dimensional image and a three-dimensional image of the machining product; the simulation execution unit 12 simulates machining by laser machining or cutting by a known analysis method such as a finite element method or a first principle calculation; the simulation execution unit 12 performs the simulation by executing, e.g., a program for a computer aided engineering (CAE); the simulation model included in the simulation execution unit includes, e.g., various calculation formulas (calculation formulas for analyzing a diameter of a machining hole, a machining depth, width of a machining groove, and the like) executed in the program for CAE, and parameters to apply to the formula; the parameters include internal parameters (parameters related to the performance of the machine tool 3 and parameters related to the material) that are set internally, in addition to external parameters that set the machining detail information and the setting condition information that are input from the outside; the simulation execution unit 12 has an inverse analysis function of setting detail information on the basis of the simulation model when the machining detail information is given; ¶ [0055] with FIG. 1: the user inputs the machining detail information and the setting condition information to the simulation device 10 before performing machining with the machine tool 3, and causes the simulation device 10 to execute the simulation; ¶¶ [0057]-[0060] with S11-S13 in FIG. 3: the input/output unit 11 receives the input of the machining detail information and the simulation execution request (step S11); the model optimization unit 14 selects the machining result information similar to the machining detail information input by the user among the machining result information accumulated in the storage unit 16, and specifies the machining detail information and the setting condition information stored in association with the selected machining result information (step S12); i.e., the model optimization unit 14 sets the specified machining detail information and setting condition information as input parameters of the simulation model; the simulation execution unit 12 sets a prescribed initial value to the internal parameters related to the performance and the like of the machine tool 3 and the internal parameters related to the material; the simulation execution unit 12 executes the machining simulation on the basis of the simulation model (step S13), and calculates the simulation result); and a machining condition evaluation unit that evaluates the machining condition based on a result of the simulation by the simulation unit (GOYA, ¶¶ [0033]-[0034] with FIG. 1: the user of the machine tool 3 confirms whether a desired machining result can be obtained under the input setting condition by inputting the machining detail and a value selected from the range of the setting condition to the simulation device 10 and referring to the machining result calculated by the simulation device 10; the user adjusts the value of the setting condition selected from the range of the setting condition until the desired machining result is obtained; when an appropriate setting condition is obtained, the user sets the setting condition in the machine tool 3 and starts actual machining on the object to be machined; in a case where the simulation by the simulation device 10 deviates from actual machining by the machine tool 3, there is a possibility that the setting condition set by the simulation device 10 is not appropriate, and the quality of the machining result by the machine tool 3 is not sufficient; to solve such a problem, the simulation device 10 has a function of adjusting various parameters of an analysis model used for the machining simulation; the various parameters are parameters related to the function and performance of the machine tool 3 or parameters related to the material of the object to be machined; the accuracy of the simulation model can be improved by adjusting the various parameters depending on actual machining by the machine tool 3 and the object to be machined, and the machining result calculated by the simulation device 10 can be closer to actual machining result; ¶¶ [0039]-[0042] with FIG. 1: the machining result evaluation unit 13 compares the machining result information acquired by the input/output unit 11 with the simulation result information calculated by the simulation execution unit 12, and evaluates the simulation result by the simulation execution unit 12; the model optimization unit 14 performs processing of optimizing the simulation performed by the simulation execution unit 12; e.g., the model optimization unit 14 optimizes the simulation by adjusting the values of the internal parameters of the simulation model on the basis of the evaluation result by the machining result evaluation unit 13; the learning unit 15 learns the values of the internal parameters optimized by the model optimization unit 14 to further improve the accuracy of the simulation model; stores the machining detail information, the setting condition information, the machining result information, the values of the internal parameters of the simulation model, and the like in machining performed by the machine tool 3; stores a large number of the machining result information received from a plurality of different machine tools such as the machine tools 3, 3a, and 3b in association with the machining detail information and the setting condition information at that time; ¶ [0055] with FIG. 1: the user adjusts the setting condition with reference to the simulation result, and repeats the operation of causing the simulation device 10 to execute the simulation again until the simulation result satisfies the request; as a result, an appropriate setting condition for certain machining detail is determined, and a mass production of the object to be machined is enabled; for that purpose, high accuracy is required for the simulation by the simulation device 10; ¶¶ [0060]-[0064] with S14-S16 in FIG. 3: the machining result evaluation unit 13 compares the machining result information selected in step S12 with the simulation result information to evaluate the degree of coincidence (step S14); in a case where the degree of coincidence of all items is equal to or more than a threshold value (step S14; YES), since the simulation result calculated by the simulation execution unit 12 is almost equal to the machining result when actually machined with machine tool 3 and the accuracy of the simulation model is sufficiently high, it is considered that the adjustment of the internal parameters is not necessary; the model optimization unit stores the currently set internal parameters (the internal parameters related to the performance and the like of the machine tool 3, the internal parameters related to the material) in the storage unit 16 in association with the machining detail information, the setting condition information, the simulation result information, and the degree of coincidence (step S16), and ends the processing of the flowchart; in a case where there are the items of which the degree of coincidence is less than a threshold value (step S14; No), the model optimization unit 14 adjusts the internal parameters (step S15); the learning unit 15 learns the items having the difference, the difference, and a relationship between the internal parameters to be adjusted and adjustment amount, and the model optimization unit 14 may adjust the parameters on the basis of the learning result; after adjusting the internal parameters, the processing from step S13 is repeated; thereafter, the simulation execution unit 12 repeatedly executes the calculation of the simulation result while changing the internal parameters until the degree of coincidence between the machining result information and the simulation result information becomes equal to or more than a threshold value; when the degree of coincidence becomes equal to or more than a threshold value, the simulation execution unit 12 stores values of the adjusted internal parameters, the machining detail information, the setting condition information, the simulation result information, and degree of coincidence in the storage unit in association with each other).
TSUNODA and GOYA are analogous art because they are from the same field of endeavor, a system and a method . Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to apply the teaching of GOYA to TSUNODA. Motivation for doing so would allow the setting condition to be efficiently set for obtaining a desired .
TSUNODA in view of GOYA fails to explicitly disclose wherein (1) the electric machine includes a wire electric discharge machine; and (2) update the machining information database by adding, to the machining information database, machining information data created.
NAKASHIMA teaches a system and a method for controlling electric machine using machine learning (NAKASHIMA, ABSTRACT and ¶ [0001]), wherein the electric machine includes a wire electric discharge machine (NAKASHIMA, ¶¶ [0002]-[0010] with FIGS. 7-8: a feed rate of a wire electrode needs to be appropriately controlled on a wire electric discharge machine such that a discharge gap between the wire electrode and a workpiece becomes constant; discharge gap is the distance between a wire electrode and a workpiece during a wire electrical discharge machining process (see FIG. 7); e.g., when a wire electrode is simply fed at a constant rate along a wire route, various factors will act thereon, causing changes such as increase and decrease in a discharge gap in the machining process; such changes in the discharge gap will lead to degradation in the machining precision; consequently, a wire electric discharge machine requires control of the feed rate so that the discharge gap is kept constant; however, it is difficult to directly detect the discharge gap; accordingly, alternative indicators are often used that is readily detected such as inter-electrode average voltage and discharge delay time; it has been known that the magnitude of inter-electrode average voltage and the length of the discharge delay time both have a correlation to the width of the discharge gap; instead of making the discharge gap constant, various modes of control are performed to ensure the constant nature of these alternative indicators; FIG. 8 is a block diagram that illustrates an example of the discharge feed control based on such alternative indicators; the deviation of the actual detected value with respect to the objective value of an indicator occurs, so as to make the deviation closer to zero, a feed rate command for the wire electrode is modified based on a predetermined control law; however, in the control as illustrated in FIG. 8, it is difficult to make the discharge gap constant in all cases; thus, in the actual machining, tuning based on an empirical rule is added to the control based on the detected value of the alternative indicator as illustrated in FIG. 8 to ensure the machining precision; it is known that it is preferable to increase the feed rate for the so-called outer corner relative to that for the linear portion; in the machining, the control law is corrected using the correction parameter, where the value of the correction parameter depends on many variables, e.g., such as programmed shape, wire diameter, workpiece thickness (P1, P2, P3); in order to carry out control that includes correction processing based on such a conventional empirical rule, it is necessary to conduct extensive/exhaustive experimentation with widely varying the variables that affects the correction parameter to determine the correction parameter; ABSTRACT and ¶¶ [0010]-[0017]: provide a control device of a wire electric discharge machine and a machine learning device that are capable of appropriately and readily determining a correction parameter; a control device optimizes a correction parameter for performing a wire electrical discharge machining process; a machine learning device configured to learn the correction parameter for performing the wire electrical discharge machining process; the machine learning device includes a state observation unit configured to observe, as a state variable, condition data indicative of a condition for performing the wire electrical discharge machining process, a determination data acquisition unit configured to acquire determination data, the determination data being indicative of the correction parameter of a case where machining precision is favorable in the wire electrical discharge machining process, and a learning unit configured to learn, using the state variable and the determination data, the correction parameter in association with the condition for performing the wire electrical discharge machining process; the learning unit includes an error calculation unit configured to calculate an error between a correlation model that derives, from the state variable and the determination data, the correction parameter for performing the wire electrical discharge machining process and a correction feature identified based on teacher data prepared in advance; and a model updating unit configured to update the correlation model such that the error is reduced; the learning unit carries out operation of the state variable and the determination data using a multilayer structure; a decision-making unit configured to display or output, based on a learning result by the learning unit, the correction parameter for performing the wire electrical discharge machining process; the learning unit learns, using the state variables and pieces of the determination data obtained for each of a plurality of the wire electric discharge machines, the correction parameter for performing the wire electrical discharge machining process on each of the wire electric discharge machines; optimizing a correction parameter for performing a wire electrical discharge machining process; ¶¶ [0030]-[0041] with FIGS. 1-2: a control device (hereafter simply referred to as "control device") 10 of a wire electric discharge machine includes a machine learning device 20 for autonomously learning a correction parameter of an optimal control law through so-called machine learning to perform feed rate control in a wire electrical discharge machining process, in other words, control of an indicator such as inter-electrode average voltage and discharge delay time. the control device 10 provided in the machine learning device 20 forms, through the learning, a model structure that represents correlation between various variables (typically programmed shape, wire diameter, workpiece thickness, etc.) for performing the wire electrical discharge machining process and the correction parameter in the wire electrical discharge machining process; a state observation unit 22 configured to observe, as a state variable S, condition data indicative of various conditions for performing the wire electrical discharge machining process; a determination data acquisition unit 24 configured to acquire determination data D indicative of the correction parameter in the case where a machining state is favorable; a learning unit 26 configured to carry out learning using the state variable S and the determination data D. the state variable S observed by the state observation unit 22 can include, e.g., a machining shape (e.g., concavity/convexity of a corner section, so-called sharp comer or arc, an angle in the case of the sharp corner, a curvature in the case of the arc, etc.) defined by a machining program, machining specifications (wire diameter, workpiece thickness, nozzle clearance, etc.); the determination data D acquired by the determination data acquisition unit 24 is the correction parameter, at the time of machining, of the case where, after the wire electrical discharge machining process has been done, an actual machining precision of the workpiece is measured, it is determined whether or not the desired machining precision has been achieved, and it is determined that the desired machining precision is actually achieved; the inspection and determination of the machining precision can be realized, e.g., by a measuring machine capable of inspecting unevenness of the workpiece surface, where the measuring machine determines whether or not the workpiece that has been machined satisfies a predetermined condition; the determination data D is a correction parameter of the case where the wire electrical discharge machining process is carried out with the state variable S and the machining state is favorable, where the determination data D indirectly represents one state of a favorable environment for performing the wire electrical discharge machining process; the learning unit 26 learns the relationship between the state variable S and the determination data D in the wire electrical discharge machining process in accordance with any learning algorithm generically called machine learning; by repeating the learning cycle of this kind, the learning unit 26 can automatically identify a feature that implies the correlation between the state variable S and the determination data D in the case where the result of the electrical discharge machining is favorable; when the correlation between the state variable S and the determination data D is interpreted to a certain reliable level, the learning results that are repeatedly output by the learning unit 26 is allowed to be used to carry out selection of an action (that is, decision making) of what kind of correction parameter should be used in the electrical discharge machining process in the current state (that is, a machining shape, machining specifications, etc.); that is, the learning unit 26, in response to the progress of the learning algorithm, can allow the correlation between the action regarding the correction parameter to be used and the current state to get gradually closer to the optimum solution; using the learning results of the learning unit 26, to obtain the optimal correction parameter in accordance with the machining shape, the machining specifications, and the like automatically and accurately without relying on experiences, complicated correspondence tables, or the like; when the optimal correction parameter in accordance with the machining shape, the machining specifications, and the like are allowed to be automatically obtained without relying on experiences or the like, the correction parameter to be applied to the control law can be quickly determined just by carrying out the analysis on the machining program before starting the wire electrical discharge machining process and entering specification data including a wire diameter, a workpiece thickness, etc.; the correlation model for estimating the desired output in response to a new input (in the machine learning device 20, the correction parameter in a case where the wire electrical discharge machining process is performed under a certain condition) is learned; an error calculation unit 32 configured to calculate an error E between a correlation model M that derives, from the state variable S and the determination data D, an optimal correction parameter for performing the wire electrical discharge machining process, and a correlation feature identified from previously prepared teacher data T, and a model updating unit 34 configured to update the correlation model M such that the error E is reduced; the learning unit 26 learns the optimal correction parameter under a given machining condition by the model updating unit 34 repeating updating of the correlation model M; the error calculation unit 32 identifies, from large quantities of teacher data T given to the learning unit 26, a correlation feature that implies the correlation between the machining shape and the machining specifications and the appropriate correction parameter, and obtains the error E between this correlation feature and the correlation model M corresponding to the state variable S and the determination data D in the current state; the model updating unit 34 updates the correlation model M, e.g., in accordance with a predefined update rule in the direction that reduces the error E; the correlation between the current state (machining shape and machining specifications) of the environment that has been unknown and an action in response thereto (decision of an appropriate correction parameter) is gradually revealed; i.e., by updating of the correlation model M, the relationship between the machining shape and the machining specifications and the appropriate correction parameter for performing the wire electrical discharge machining process with the machining shape and the machining specifications is gradually made closer to the optimum solution; ¶¶ [0049]-[0050] with FIG. 9: the control device of the wire electric discharge machine changes the control law while machining the linear portion and while machining the comer portion; specifically, the proportional gain is changed by giving the value of the proportional gain as K=K0+P×K1 and changing the parameter P; i.e., the parameter P is the sensitivity to the change in the proportional gain; the control device of the wire electric discharge machine carries out the machining of the linear portion with K=K0, where K0 can be determined as appropriate using various prior art techniques; meanwhile, machining of the comer portion is performed with K=K0+P×K1, where the control device of the wire electric discharge machine inputs, as the state variable S, the various conditions of the electrical discharge machining which is currently to be performed to the control device 10, and thereby obtains the determination data D corresponding to the state variable S, i.e., the correction parameter; by using this correction parameter as P, the optimal gain can be obtained; ¶¶ [0051]-[0055] with FIG. 4: the decision-making unit 52 displays the optimal correction parameter for performing the wire electrical discharge machining process learned by the learning unit 26 for an operator, or generates and outputs the command value C for the wire electric discharge machine based on the optimal correction parameter for performing the wire electrical discharge machining process that has been learned by the learning unit 26; when the machining state of the workpiece machined using the correction parameter displayed or output to the environment by the decision-making unit 52 is favorable, then the determination data acquisition unit 24 acquires the correction parameter as the determination data D in the next learning cycle; the learning unit 26 continues the learning using the state variable S and the determination data D that have been input, e.g., by updating the correlation model M; the machine learning device 50 can change the state of the environment by the output of the decision-making unit 52).
TSUNOD and NAKASHIMA are analogous art because they are from the same field of endeavor, a system and a method for controlling electric machine using machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to apply the teaching of NAKASHIMA to TSUNODA in view of GOYA. Motivation for doing so would expand the capability of controlling various electric machine tools for machining various materials.
TSUNODA in view of GOYA and NAKASHIMA fails to explicitly disclose update the machining information database by adding, to the machining information database, machining information data created.
Akemura teaches a system and a method relating to an electrical discharge machine ("EDM"), wherein update the machining information database by adding, to the machining information database, machining information data created (Akemura, Col. 1, line 15 – Col. 1line 14: in electrical discharge machining, the work piece is electrical-discharge-machined by applying the voltage pulse to a clearance, called the gap, between the work piece and the tool electrode (hereinafter the "electrode") while the electrode is fed in the vertical direction relative to the workpiece; a cavity having the desired dimensions and surface roughness is formed in the workpiece by means of the voltage pulse applied to the machining gap based on machining condition parameters including a certain OFF time of the voltage pulse applied to the gap and a certain ON time of the current pulse flowing through the gap, and the peak current; the operator, when preparing for electrical discharge machining, first determines the workpiece material, and the dimensions and material of one or more electrodes to be used in accordance with the desired dimensions and the desired surface roughness of the cavity to be formed to the workpiece; a plurality of sets of machining conditions, including the current ON time, the peak current, and the OFF time of the voltage pulse applied to the gap are determined, and a machining program is compiled; the machining conditions include the feed of the electrode relative to the workpiece; further, in applications where the workpiece is machined using a single electrode, the electrode may move relative to the workpiece in a plane perpendicular to the direction in which the electrode is vertically fed; the machining conditions may include the amplitude of this movement of the electrode; determining these machining conditions require a high degree of skill; to this end, an automatic programming device is proposed that compiles a machining program by automatically determining a plurality of sets of machining conditions in accordance with setting data from specifications including the desired workpiece or cavity dimensions, the desired surface roughness, the workpiece material, and the dimensions and material of the electrode to be used; such an automatic programming device stores a data table which includes the relationship between the plurality of sets of the machining conditions and machining results including the surface roughness obtained by the use of those machining conditions; from these plural sets of machining conditions, the set which will yield machining results which are approximate to the specifications are selected; a method of calculating machining conditions which are identical with or approximate to the specifications has been considered using a functional equation derived by expressing the relationship between the specifications, including such factors as the desired Surface roughness, machining area, and machining depth; and machining conditions including the peak current; provide an electrical discharge machine that determines the machining conditions most suited to the machining specifications including the desired surface roughness; Col. 2, lines 18-67: a method of setting machining conditions for electrical discharge machining is provided that sets the machining conditions, including the peak current and the feed of the electrode, based on input specifications including the desired surface roughness; setting data related to certain specifications, including the material of the workpiece and the desired surface roughness of the product (via an input unit), storing a plurality of sets of basic data indicative of the relationship between the specifications and the machining conditions (via a basic data storage unit), selecting plural sets of peripheral data having specifications most suited to the set specifications from the plurality of sets of basic data (via data reader), studying the relationship between the specifications and the machining conditions using the plural sets of peripheral data (via an inference unit), and inferring and compiling data for determining machining conditions suited to said input specifications and the relative movement of the electrode and the workpiece based on said study (via an inference control unit); adding the inferred and compiled data to the basic data (stored in the basic data storage unit), and using the basic data, including the added inferred and compiled data, when performing a subsequent step of inferring; Col. 5, line 7 – Col. 10, line 64 with FIGS. 1 and 5-9: the "IP value" is the set value of the peak current corresponds to a setting representing a peak current of about 1.5 A per 1 IP; the inference control unit 46 commands the inference unit 43 to study the relationship between each element of peripheral data, based on the peripheral data groups D1 to D8 thus selected; the inference unit 43, using the neural network model shown in FIG. 6, studies the peripheral data groups, D1 to D8 selected from the basic data storage unit 42; the machining area, machining depth, and reduction in electrode dimension in the above-mentioned data groups D1 to D8 are applied to the input side of the neural network model, and the IP value is applied to the output side; inference will be done using the actually input machining area (80 mm), machining depth (15 mm), and reduction in electrode dimension (360 pm) as input signals to obtain the most suitable IP α value; the steps of studying and inferring here preferably use a neural network such as that shown in FIG. 6 study and create inferences using the above-mentioned input data; after studying the relationship between the machining area, the surface roughness corresponding to each IP value, and the IP value (see step S42 in FIG. 9), the inference control unit 46 causes the inference unit 43 to infer an IP value o(2) which will yield about half of the surface roughness obtained with the first IP value using data corresponding to the actual machining area and a surface rough ness of d/2"; the inference control unit 46 compares the values d/2" to the required surface roughness, and repeats the above-mentioned steps S42 and S43 of FIG. 9 until d/2" becomes smaller than the required surface roughness; a plurality of peak current values IP α(1) to (n), are inferred and stored in the momentary storage unit 44 as the second, the third, so on, and the nth peak current (see step S44 in FIG. 9); FIG. 8 is a table showing the relationship regarding the study data and inference data used for inferring and compiling the peak current values up to the nth and the results; all necessary machining conditions, including the amplitude and the feed of the electrode at each machining stage, are calculated from the obtained values of α, β, ϵ and ζ; for each value of the IP α(1), reduction in electrode dimension β, side face residue ϵ, etc. obtained as mentioned above is stored in the momentary storage unit 44, and when the results of the machining with the machining conditioning data (SD) are excellent, the operator issues an instruction to retain the machining conditions; the inferred data (FD) such as the data IP value, surface roughness, bottom gap, etc. used is also accumulated in the preliminary data storage unit, inside the basic data storage unit 42, as a part of the basic data along with the input values such as the machining area, machining depth, and the material(s); in the inferred data is stored so as to be able to be used as one of the basic data at the inference stage when machining is done using a subsequent machining specification; in other words, when inferring, the data reader 41 first seeks the proximate value from the stored basic data, and then, the available database itself, progress so that the inference may be done with better accuracy by adding more proximate data to the preliminary data storage unit, to be used as study data).
TSUNODA in view of GOYA and NAKASHIMA, and Akemura are analogous art because they are from the same field of endeavor, a system and a method relating to an electrical discharge machine ("EDM"). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to apply the teaching of Akemura to TSUNODA in view of GOYA and NAKASHIMA. Motivation for doing so would provide better machining accuracy (TSUNODA, Col. 10, lines 44-64).
Claim 2
TSUNODA in view of GOYA, NAKASHIMA, and Akemura discloses all the elements as stated in Claim 1 and further discloses learns, based on the machining information data stored in the machining information database, a correlation among the data related to the machining content, the data related to the required specification, and the data related to the machining condition (NAKASHIMA, ¶¶ [0030]-[0041] with FIGS. 1-2: a control device (hereafter simply referred to as "control device") 10 of a wire electric discharge machine includes a machine learning device 20 for autonomously learning a correction parameter of an optimal control law through so-called machine learning to perform feed rate control in a wire electrical discharge machining process, in other words, control of an indicator such as inter-electrode average voltage and discharge delay time. the control device 10 provided in the machine learning device 20 forms, through the learning, a model structure that represents correlation between various variables (typically programmed shape, wire diameter, workpiece thickness, etc.) for performing the wire electrical discharge machining process and the correction parameter in the wire electrical discharge machining process; a state observation unit 22 configured to observe, as a state variable S, condition data indicative of various conditions for performing the wire electrical discharge machining process; a determination data acquisition unit 24 configured to acquire determination data D indicative of the correction parameter in the case where a machining state is favorable; a learning unit 26 configured to carry out learning using the state variable S and the determination data D. the state variable S observed by the state observation unit 22 can include, e.g., a machining shape (e.g., concavity/convexity of a corner section, so-called sharp comer or arc, an angle in the case of the sharp corner, a curvature in the case of the arc, etc.) defined by a machining program, machining specifications (wire diameter, workpiece thickness, nozzle clearance, etc.); the determination data D acquired by the determination data acquisition unit 24 is the correction parameter, at the time of machining, of the case where, after the wire electrical discharge machining process has been done, an actual machining precision of the workpiece is measured, it is determined whether or not the desired machining precision has been achieved, and it is determined that the desired machining precision is actually achieved; the inspection and determination of the machining precision can be realized, e.g., by a measuring machine capable of inspecting unevenness of the workpiece surface, where the measuring machine determines whether or not the workpiece that has been machined satisfies a predetermined condition; the determination data D is a correction parameter of the case where the wire electrical discharge machining process is carried out with the state variable S and the machining state is favorable, where the determination data D indirectly represents one state of a favorable environment for performing the wire electrical discharge machining process; the learning unit 26 learns the relationship between the state variable S and the determination data D in the wire electrical discharge machining process in accordance with any learning algorithm generically called machine learning; by repeating the learning cycle of this kind, the learning unit 26 can automatically identify a feature that implies the correlation between the state variable S and the determination data D in the case where the result of the electrical discharge machining is favorable; when the correlation between the state variable S and the determination data D is interpreted to a certain reliable level, the learning results that are repeatedly output by the learning unit 26 is allowed to be used to carry out selection of an action (that is, decision making) of what kind of correction parameter should be used in the electrical discharge machining process in the current state (that is, a machining shape, machining specifications, etc.); that is, the learning unit 26, in response to the progress of the learning algorithm, can allow the correlation between the action regarding the correction parameter to be used and the current state to get gradually closer to the optimum solution; using the learning results of the learning unit 26, to obtain the optimal correction parameter in accordance with the machining shape, the machining specifications, and the like automatically and accurately without relying on experiences, complicated correspondence tables, or the like; when the optimal correction parameter in accordance with the machining shape, the machining specifications, and the like are allowed to be automatically obtained without relying on experiences or the like, the correction parameter to be applied to the control law can be quickly determined just by carrying out the analysis on the machining program before starting the wire electrical discharge machining process and entering specification data including a wire diameter, a workpiece thickness, etc.; the correlation model for estimating the desired output in response to a new input (in the machine learning device 20, the correction parameter in a case where the wire electrical discharge machining process is performed under a certain condition) is learned; an error calculation unit 32 configured to calculate an error E between a correlation model M that derives, from the state variable S and the determination data D, an optimal correction parameter for performing the wire electrical discharge machining process, and a correlation feature identified from previously prepared teacher data T, and a model updating unit 34 configured to update the correlation model M such that the error E is reduced; the learning unit 26 learns the optimal correction parameter under a given machining condition by the model updating unit 34 repeating updating of the correlation model M; the error calculation unit 32 identifies, from large quantities of teacher data T given to the learning unit 26, a correlation feature that implies the correlation between the machining shape and the machining specifications and the appropriate correction parameter, and obtains the error E between this correlation feature and the correlation model M corresponding to the state variable S and the determination data D in the current state; the model updating unit 34 updates the correlation model M, e.g., in accordance with a predefined update rule in the direction that reduces the error E; the correlation between the current state (machining shape and machining specifications) of the environment that has been unknown and an action in response thereto (decision of an appropriate correction parameter) is gradually revealed; i.e., by updating of the correlation model M, the relationship between the machining shape and the machining specifications and the appropriate correction parameter for performing the wire electrical discharge machining process with the machining shape and the machining specifications is gradually made closer to the optimum solution; ¶¶ [0051]-[0055] with FIG. 4: the decision-making unit 52 displays the optimal correction parameter for performing the wire electrical discharge machining process learned by the learning unit 26 for an operator, or generates and outputs the command value C for the wire electric discharge machine based on the optimal correction parameter for performing the wire electrical discharge machining process that has been learned by the learning unit 26; when the machining state of the workpiece machined using the correction parameter displayed or output to the environment by the decision-making unit 52 is favorable, then the determination data acquisition unit 24 acquires the correction parameter as the determination data D in the next learning cycle; the learning unit 26 continues the learning using the state variable S and the determination data D that have been input, e.g., by updating the correlation model M; the machine learning device 50 can change the state of the environment by the output of the decision-making unit 52); and derive, based on a learning result of the correlation, data related to the at least one machining condition estimated to satisfy the desired machining content and the required specifications (TSUNODA, ¶¶ [0006]-[0007]: adjust at least either of machining conditions and machining parameters in consideration of particulars demanded in accordance with a machining type of a workpiece in a machine tool; ¶ [0032] wit FIG. 1: the machine learning device 100 is capable of observing information (the information on the tools such as the types of the tools, the information on the cutting conditions such as the spindle speed, the feed speed, and the cutting depth, the information; on the workpieces such as the materials and the shapes of the workpieces, the power to be consumed by each motor, the dimension values and the surface quality of the portions of the machined workpieces and the temperatures of the portions of the machine tool that have been measured by the sensors 3, or the like, for instance) that may be acquired in the machining condition adjustment device 1, through the interface 21; ¶¶ [0048] and [0053] with FIG.2: the learning model selection unit 105 selects a learning model corresponding to the machining type inputted from the preprocessing unit 36 and causes the selected learning model to be used for the learning by the learning unit 110 and decision making by the decision making unit 122; the learning model may be configured so as to carry out more effective learning and inferencing by using a so-called deep learning technique with use of a neural network that forms three or more layers; the learning model is used for determination of the adjustment behavior for at least either of the machining conditions and the machining parameters by the decision making unit 122; ¶¶ [0068], [0070], [0072], and [0075]-[0077] with FIGS. 4-5: the learning model generated by the learning unit 110 may be used for estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to the machining type, when the state of machining by the machine tool 2 is given; generates a plurality of learning models in which at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and the state of machining by the machine tool 2 have been learned in association; with use of the plurality of learning models generated in this manner, an estimation unit 120 is capable of carrying out estimation processing that is based on the state data S acquired from the machine tool 2 and that is demanded for determination of at least either of the machining conditions and the machining parameters which are more appropriate and which correspond to the machining type in the acquired state; the machining condition adjustment device 1 includes a configuration for estimation that is demanded when the machine learning device 100 estimates at least either of machining conditions and machining parameters for machining by each machine tool; in a stage of the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model by the machine learning device 100, the preprocessing unit 36 carries out the conversion (such as digitization or sampling) into the unified format that is handled in the machine learning device 100, based on the data acquired by the data acquisition unit 34 and the priority condition data stored in the priority condition storage unit 56; the state data S in a given format that is used for the estimation by the machine learning device 100 is produced from the converted data and the produced state data S, together with the machining type, is outputted to the machine learning device 100; the estimation unit 120 carries out the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model selected by the learning model selection unit 105, based on the state data S produced by the preprocessing unit 36; in the estimation unit 120, the state data S inputted from the preprocessing unit 36 is inputted into the learning model generated (having the parameters determined) by the learning unit 110 and at least either of the machining conditions and the machining parameters satisfying the priority condition associated with the machining type are thereby estimated and outputted; the machining condition adjustment device 1 is capable of estimating at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and controlling the machining operation for a workpiece by the machine tool 2 based on at least either of the machining conditions and the machining parameters that have been estimated; ¶¶ [0089], [0091], and [0093]-[0097] with FIGS. 6-7: the learning model generated by the learning unit 110 may be used for estimation of at least either of the machining conditions and the machining parameters satisfying the priorities corresponding to the machining type in the machining by the machine tool 2; generates a plurality of learning models in which the distribution of at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type has been learned. With use of the plurality of learning models generated in this manner, the estimation unit 120 is capable of carrying out estimation processing that is based on the state data S acquired from the machine tool 2 and that is demanded for determination of at least either of the machining conditions and the machining parameters which are more appropriate and which correspond to the machining type in the acquired state; the machining condition adjustment device 1 includes a configuration that is demanded when the machine learning device 100 estimates at least either of machining conditions and machining parameters for machining by each machine tool; in a stage of the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model by the machine learning device 100, the preprocessing unit 36 carries out the conversion (such as digitization or sampling) into the unified format that is handled in the machine learning device 100, based on the data acquired by the data acquisition unit 34 and the priority condition data stored in the priority condition storage unit 56; the state data S in a given format that is used for the estimation by the machine learning device 100 is produced from the converted data and the produced state data S, together with the machining type, is outputted to the machine learning device 100; the estimation unit 120 carries out the estimation of at least either of the machining conditions and the machining parameters satisfying the priority condition with use of the learning model selected by the learning model selection unit 105, based on the state data S produced by the preprocessing unit 36; in the estimation unit 120 of the embodiment, at least either of the machining conditions and the machining parameters satisfying the priority condition are estimated and outputted based on a position of the state data S inputted from the preprocessing unit 36 in a distribution of data included in the learning model generated by the learning unit 110; ¶¶ [0042] and [0049]-[0052] with FIG. 2: on condition that the machine learning device 100 carries out the reinforcement learning, e.g., the preprocessing unit 36 produces a set of state data S and determination data D in given formats in the learning, as the learning data; the reinforcement learning is the technique in which a cycle including observing a current state (that is, input) of an environment where a learning object exists, executing given behavior (that is, output) in the current state, and conferring some reward for the behavior is iterated by a trial-and-error method and in which a measure (the adjustment behavior for at least either of the machining conditions and the machining parameters in the machine learning device 100 of the application) that maximizes total of such rewards is learned as an optimal solution; in the Q-learning by the learning unit 110, the reward R may be determined based on the priority condition associated with the machining type stored in the priority condition storage unit 56; the state data S which is the current learning object is based on the data acquired during the drilling; the reward R may be positive (plus) when the pitch error as the determination data is smaller than the threshold Errpit stored in the priority condition storage unit 56 or may be negative (minus) when the pitch error is equal to or greater than the threshold Errpit; the pitch error magnitude of the positive or negative reward may be changed in accordance with a deviation of the pitch error from the threshold Errpit; the reward R may be calculated with use of an expression in which the plurality of priority conditions are combined; the state data S which is the current learning object is based on the data acquired during the finishing, e.g., a given formula for reward calculation in which data related to the shape precision and data related to the cycle time are used may be defined in advance and the reward R may be calculated with use of the formula for reward calculation; the learning unit 110 may use a neural network as a value function Q (learning model) and may be configured so as to use the state data S and behavior a as input of the neural network and so as to output a value (result y) of the behavior a in a pertinent state; ¶¶ [0055]-[0056]: the decision making unit 122 determines the optimal solution of the adjustment behavior for at least either of the machining conditions and the machining parameters with use of the learning model selected by the learning model selection unit 105 based on the state data S inputted from the preprocessing unit 36 and outputs the determined adjustment behavior for at least either of the machining conditions and the machining parameters; the decision making unit 122 inputs the state data S (the tool data S1, the machining condition data S2, and the machining parameter data S3) inputted from the preprocessing unit 36 and the adjustment behavior for at least either of the machining conditions and the machining parameters (a combination of adjustment for the feed speed, adjustment for the spindle speed, and the like, or change in setting of the parameters) as input data into the learned model updated (having the parameters determined) through the reinforcement learning by the learning unit 110, so that the reward in case where the pertinent behavior is executed in the current state is calculated; the calculation of the reward in the decision making unit 122 is carried out for the adjustment behavior for at least either of the machining conditions and the machining parameters that may be currently adopted; through a comparison among a plurality of calculated rewards, the adjustment behavior for at least either of the machining conditions and the machining parameters that results in calculation of the largest reward is determined as the optimal solution; the optimal solution of the adjustment behavior for at least either of the machining conditions and the machining parameters determined by decision making unit 122 is inputted into the control unit 32 and is used for determination of at least either of the machining conditions and the machining parameters in actual machining; the machining condition adjustment device 1 can adjust appropriately for at least either of the machining conditions and the machining parameters in accordance with particulars demanded by an operator during the machining of a workpiece by the machine tool 2; ¶¶ [0098]-[0101] with FIG. 7: the estimation unit 120 calculates a distance between each data set (cluster) in the distribution of the data included in the learning model generated by the learning unit 110 and the position of the state data S inputted from the preprocessing unit 36 and estimates that the current machining conditions or machining parameters satisfy the priority condition in the current machining type, in case where the distance between the position of the state data S inputted from the preprocessing unit 36 and a nearest data set C1n is equal to or shorter than a predetermined and given threshold Distth1, for instance; the estimation unit 120 estimates that the priority condition in the current machining type is not satisfied because of at least either of the current machining conditions and machining parameters, in case where the distance between the position of the state data S inputted from the preprocessing unit 36 and the nearest data set C1n is longer than the predetermined and given threshold Distth1, for instance; the estimation unit 120 adjusts one machining condition or machining parameter or a plurality of machining conditions or machining parameters within the state data S inputted from the preprocessing unit 36 in accordance with a predetermined and given rule so that the distance to the data set C1n may be made equal to or shorter than the predetermined and given threshold Distth1; the predetermined rule for such adjustment may be a rule by which a given machining condition or a given machining parameter is fixedly adjusted, for instance; the rule may provide that the adjustment shall be made so that the distance to the data set C1n may be made equal to or shorter than the predetermined and given threshold Distth1 by a smallest adjustment amount; there may be a rule that a given machining condition or a given machining parameter shall be excluded from objects of the adjustment; thus the estimation unit 120 estimates and outputs at least either of the machining conditions and the machining parameters satisfying the priority condition, based on the current state data S and the learning model; the estimation unit 120 may output an instruction to stop the machining, in case where the distance between the position of the state data S inputted from the preprocessing unit 36 and the nearest data set C1n is longer than a predetermined and given threshold Distth2 (threshold Distth2 > threshold Distth1); an instruction for emergency stop may be outputted in case where the distance is longer than a predetermined and given threshold Distth3 (threshold Distth3 > threshold Distth2); such a technique makes it possible to estimate that the state of machining is abnormal, in case where operation vastly different from normal operation is carried out, and to call attention of the operator; the machining condition adjustment device 1 is capable of estimating at least either of the machining conditions and the machining parameters satisfying the priority condition corresponding to each machining type and controlling the machining operation for a workpiece by the machine tool 2 based on at least either of the machining conditions and the machining parameters that have been estimated).
Response to Arguments
Applicant’s arguments filed on 06/29/2026 with respect to Claims 1-2 have been fully considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Sekiai et al. (US 2008/0168016 A1, pub. Date: 07/10/2008) discloses in ¶ [0019] that provide a plant control apparatus having operation signal generating unit for calculating an operation signal to be supplied to a plant by using a measurement signal representative of a running state amount of the plant, comprising: (a) a measurement signal database for storing a past measurement signal; (b) an operation signal database for storing a past operation signal; (c) numerical value analysis executing unit for analyzing running characteristics of the plant; (d) a numerical value analysis result database for storing numerical analysis results obtained by operating the numerical analysis executing unit; (e) a model for estimating a value of a measurement signal obtained when an operation signal is applied to the plant, by using information in the numerical value analysis result database; (f) learning unit for learning a plant operation method by using the model; (g) a learning information database for storing learning information obtained by the learning unit; (h) a control logic database for storing information to be used by the operation signal generating unit when an operation signal is derived; (i) a knowledge database for storing knowledge regarding the running characteristics of the plant; (j) analyzing unit for processing information in the numerical value analysis result database, by using information in said knowledge database, the learning information database, the operation signal database and the measurement signal database; and (k) an analysis result database for storing an analysis result by the analyzing unit, wherein the analyzing unit includes at least one of learning base analyzing unit for evaluating adequacy of the operation method learnt by said learning unit, signal analyzing unit for evaluating an effect obtained if the operation signal is applied to the plant and presence/absence of an abnormal measurement signal, and knowledge database updating unit for adding information to or correcting information in, said knowledge database. Sekiai further discloses in ¶ [0103] that (1) the knowledge database updating unit 330 evaluates know ledge in the know ledge database 400 by referring to the information in the numerical value analysis result database 240, measurement signal database 210, and analysis result database 500; and (2) when necessary, the knowledge database updating unit is provided with a function of correcting knowledge in the knowledge database 400 or a function of adding new knowledge. Sekiai also discloses in ¶¶ [0117]-[0126] with FIG. 5 that (1) the knowledge database updating unit 330 has three functional blocks including a knowledge evaluation functional block, a knowledge correction functional block and a knowledge addition functional block; (2) Step 2300 selects one of three blocks to be used; (3) the knowledge evaluation functional block evaluates validity of the knowledge in the knowledge database 400; (4) at Step 2310 the knowledge used by the learning base analyzing unit 310 and signal analyzing unit 320 is extracted from the analysis result database 500, and drives the number of use times of each knowledge in the knowledge database 400; (5) in accordance with the number of use times derived at Step 2310, Step 2320 calculates an evaluation value for each knowledge; e.g., an evaluation value of the knowledge having a larger number of use times is set to a large value, whereas an evaluation value of the knowledge having a smaller number of use times is set to a small value; (6) Step 2330 transmits the evaluation value calculated at Step 2330 to the knowledge database 400 to be stored therein; (7) the knowledge correction functional block compares the information in the measurement signal database 210, operation signal database 250 and numerical value analysis result database 240 with the knowledge in the knowledge database 400; (8) if the knowledge in the knowledge database 400 contradicts the information in the measurement signal database 210, operation signal database 250 and numerical value analysis result database 240, it is judged that the knowledge has an error, and the knowledge in the knowledge database 400 is corrected; (9) in accordance with this judgment, Step 2410 selects knowledge to be corrected; (10) Step 2420 corrects the knowledge so as to eliminate the contradiction; (11) Step 2430 transmits the new knowledge corrected at Step 2420 to the knowledge database 400 to be stored therein.; (12) the knowledge addition functional block generates a hypothesis capable of explaining one numerical value analysis result in the numerical value analysis result database 240; (13) if this hypothesis can explain sufficiently other numerical value analysis results, this hypothesis is registered as knowledge in the knowledge database 400; (14) at Step 2510 a numerical value analysis result is extracted from the numerical value analysis result database 240; (15) Step 2520 generates a hypothesis capable of explaining the numerical value analysis result extracted at Step 2510; (17) it is evaluated at Step 2530 whether the hypothesis generated at Step 2520 can explain other numerical value analysis results in the numerical value analysis result database 240; (18) it is judged at Step 2540 from the evaluation results at Step 2530 whether the hypothesis is to be added to the knowledge database 400; (19) if the hypothesis is to be added, the flow advances to Step 2550, whereas not, the knowledge addition functional block is terminated; and (20) at Step 2550 the hypothesis generated at Step 2530 is transmitted to the knowledge database 400 to be stored therein.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HWEI-MIN LU whose telephone number is (313)446-4913. The examiner can normally be reached Mon - Fri: 9:00 AM - 6:00 PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Mariela D. Reyes can be reached at (571) 270-1006. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HWEI-MIN LU/Primary Examiner, Art Unit 2142
Prosecution Insights