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 Arguments
Applicant’s response related to claim interpretation is acknowledged.
Applicant’s argument regarding claim 1 that neither in combination nor individually JP34 and CN82 teach predicting a load parameter value of an electrical equipment for a future time based on at least one machine learning model and a plurality of load parameter values because an estimated accident is not load parameter value have been fully considered but not found persuasive. In JP34, page 11, 1st paragraph the machine learning model determines the estimated accident for the future time based on predicted load parameter of the transmission line connected to transformer value in addition to other state values for the future time such as the next 30 minutes as taught in page 9, 2nd-3rd paragraph and page 12, 2nd paragraph. The future system state predicts the power demand for the next 30 minutes which is the load parameter of the transmission line and based on that information and other information an accident is estimated if the load exceeds the overload margin of the transmission including the load parameter of the transmission line. How would the system know there would be an accident based on energy fluctuation if the model does not know about the load on the transformer and the transmission line for the future time period. As such in JP34 teaches predicting a load parameter value of an electrical equipment for a future time based on at least one machine learning model.
Applicant further argued neither in combination nor individually JP34 and CN82 teaches to determine overload capacity overload capacity value for the electrical equipment for the future time based on the predicted load parameter value. Examiner did not find the argument fully persuasive. In the response applicant conceded that JP34 do teaches setting overload capacity for a current duration but not for future but applicant did not specify the future time does not include from the current time till a certain period. In other words, future time could be for this time period till the next 30 minutes which is clearly taught in JP34 page 12, 2nd paragraph. Applicant also argued that overload capacity to a time is not calculating overload capacity value but did not explain why the overload capacity value in the claim is different than overload capacity to a time. In view of [0054] of the specification, examiner understands that applicant is indicating overload capacity value actually means percentage indicative of an amount of additional load that can be withstood by a particular electrical apparatus for a particular amount of time without abnormally and/or extensively aging or damaging the electrical equipment. However the claim does not define overload capacity value as above. Overload capacity value is a broad term in the claim which can also include overload capacity to a time. Based on applicant’s argument and the specification, for enhanced prosecution, examiner performed an updated search and found new prior art, Raith et al. which clearly teaches in [0046] and [0047], based on transformer current load value and other conditions, an overload capacity for the transformer for the next 2-8 hours or desired time period is calculated using a load forecasting model. Therefore it would have been obvious before the effective filing date of the claimed invention to a person of ordinary skill in the art to modify the method of using machine learning model to predict load parameter value of electrical equipment for future time while considering overload capability of the equipment and changing one parameter based predicted load and overload as taught by combination of JP34 and CN82 by applying the known technique calculating overload capacity value for a future or desired time as taught by Raith et al. and as an improvement to the method of calculating overload capability for future time to yield predictable results of managing overload condition in future time while considering equipment’s condition and energy supply network condition as taught by Raith et al. in [0005].
Claim Rejections - 35 USC § 103
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.
Claim(s) 1,5,6-10,13,15,16,17,19 and 20 rejected under 35 U.S.C.103 as being
unpatentable over JP34 (JP 2019216534 A) in view of Raith et al. (US 20200312538 A1) and CN82 (CN 108664682 A).
Regarding claim 1 JP34 teaches, a method (power system monitoring system
page 3, 3rd paragraph) comprising:
predicting, by a processor circuit (processing unit having different unit such as
system state unit for prediction which are realized by a processor such as CPU, page 7,
3rd paragraph), a load parameter value of an electrical equipment for a future time
based on at least machine learning model (model created by machine learning predicts or estimates an accident in the future based on future state estimation including load on the transformer in addition to energy fluctuation1 in the future time (load parameter of the equipment), page 11, 1st paragraph, page 9, 2nd-3rd paragraph and page 12, 2nd paragraph),
page 11, 1st paragraph) and a plurality of load parameter values comprising a set of predefined number of load parameter values obtained for the electrical equipment (based on current measurements of the power system (plurality of load parameter values), the plurality of predicted values of fluctuation range in energy (a load
parameter) is predicted for a power system equipment, page 2, 1st paragraph, page 3,
3rd paragraph, page 11, 3rd paragraph);
calculating, by the processor circuit, an overload capability based on the predicted load parameter value (based on the estimation result of the predicted values of fluctuation range in energy, the system predicts an accident or accidents such as transformer overload/failure might occur in future in view of determined overload tolerance for a certain amount of time as taught in page 12, 2nd paragraph, page 3, 3rd paragraph, page 5, 4th paragraph and page 15, 3rd paragraph); and
changing, by the processor circuit, at least one parameter associated with
the electrical equipment at the present time based on the calculated overload
capability for the future time (based on the predicted future accident such as
transformer overload and exceeding overload margin, the control content determination
unit determines control content/command for improving reliability (transformer
performing safely during overload) and sends the control command with control amount
to one of the selected controllable devices of the power system. The selected control
device executes the control command, page 14, 5th paragraph and page 15, 1st
paragraph).
JP34 does not teach the details of at least one overload capacity value for the electrical equipment for the future time and details of time series data. The system determine for how long the transformer can operate in a overload state based predicted load change but the exact overload capacity value is not taught. It is also not clear the system measurements considered and the predicted states include time series data or not.
Raith et al. teaches, at least one overload capacity value for the electrical equipment for the future time based, (determining overload capacity of the transformer to be operated in the next 2 hours (future time) based on determined and derived state conditions of the transformer, [0007], [0041], [0046] and [0048]2).
Therefore it would have been obvious before the effective filing date of the claimed invention to a person of ordinary skill in the art to modify the method of using machine learning model to predict load parameter value of electrical equipment for future time while considering overload capability of the equipment and changing one parameter based predicted load and overload as taught by JP34 by applying the known technique of calculating overload capacity value for a future or desired time as taught by Raith et al. and as an improvement to the method of calculating overload capability for future time to yield predictable results of managing overload condition in future time while considering equipment’s condition and energy supply network condition as taught by Raith et al. in [0005].
Neither in combination nor individually JP34 and Raith et al. teach the details of time series data.
CN82 teaches, extracted from a time series data stream of load parameter
values (for forecasting load, historical load data over an interval (time series load
parameter) is collected, page 11, 2nd and 3rd paragraph).
Therefore it would have been obvious before the effective filing date of the
claimed invention to a person of ordinary skill in the art to modify the method of predicting load parameter value and determining overload capacity of an electrical equipment for future time as taught by combination of JP34 and CN82 by applying the known technique of extracting predefined number of load parameters values from a time series data such as historical load data as taught by CN82 as an improvement to load
prediction to yield predictable results for predicting load and overload conditions.
JP34 teach:
(page 9, 2nd-3rd paragraph) The system state estimating unit 14C may acquire the measurement information from the system information collecting unit 14B or may read
the measurement information from the storage unit 13. When estimating the future system state, the system state estimating unit 14C further refers to a demand forecast related to power demand3 (demand forecast information 13C), an operation plan related to power supply in the power system 21 (operation plan information 13D), and the like. .
The future system information includes, for example, node data and branch data in each system state case i. FIG. 5 is a diagram illustrating an example of the content of the node data. FIG. 6 is a diagram illustrating an example of the content of the branch data. The future system information includes information recorded in table data in the format shown in FIGS. 5 and 6 for each system state case i. The node data shown in FIG. 5 includes a voltage, a phase, a generator output (active power output, reactive power
output), a load (active power load, reactive power load), and a The information is information associated with the information. The branch data shown in FIG. 6 is information in which active power flow, reactive power flow, active power loss, and information on reactive power loss are associated with a branch name. The node data and the branch data may include data other than the information shown in FIGS. 5 and 6, or may reduce some of the data included in the information shown in FIGS. Is also
good.
(page 12, 2nd paragraph) For example, the limit value of the overload margin is set according to the relationship between the overload capacity of the transmission line and
the transformer and the elapsed time. FIG. 7 is a reference diagram showing the relationship between the overload tolerance and time in a graph format. The overload tolerance4 is an example of a limit value of an overload margin. For example, as shown in FIG. 7, when the elapsed time is a continuous capacity, the overload tolerance is set to the continuous tolerance. When the overload has a capacity of 30 minutes, the overload capacity is set to 30 minutes capacity, and when the overload is 10 minutes capacity, the overload capacity is set to 10 minutes capacity. That is, the limit value of the overload margin is set so that the power system 21 is operated so that the power flow flowing through the transmission line does not exceed the overload withstand capability for each time zone.
Raith et al. teach:
[0046] FIG. 2 shows an example embodiment of the method 1 according to the invention which is shown schematically in FIG. 2. A load-forecasting model 2 is shown which receives a load forecast request 3 at a request time. The load forecast request 3 contains the query concerning the amount of overload with which the transformer can be operated in 2 hours5 for 8 hours if the lifetime consumption is intended to be 110%. Reference is made to the lifetime consumption (100%) which occurs during an operation of the transformer at nominal power.
[0048] According to the invention, the already consumed lifetime is in no way roughly estimated. Instead, the consumed lifetime is determined continuously on the basis of measured values and is stored on a storage unit 6. According to the invention, it is possible for the load-forecasting model 2 to determine the overload capacity of the transformer more precisely on the basis of the lifetime measured in this way or, in other words, the lifetime consumption measured in this way.
Regarding claim 5 combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition JP34 teaches, wherein the plurality of load parameter values comprises a set of at least five load parameter values iteratively extracted from a stream of load parameter values obtained from the electrical equipment (5 patterns of the
predicted average value (load parameters variation over time) of are used to determine
energy output predicted values for the power equipment, page 24, 2nd paragraph).
Regarding claim 6 combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition JP34 teaches, wherein the future time for the predicted load parameter value is at least one hour after the predicting (the power system monitoring system estimates future system state for example 60 mins (one hour) after the current time, page 3, 3rd paragraph).
Regarding claim 7 combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition JP34 teaches, wherein the at least one machine learning model predicts the load parameter value of the electrical equipment for the future time based only on the plurality of load parameter values (only power measurement values over time are used by the machine learning model to perform load forecast in future,1st paragraph, page 8, 4th paragraph, page 4, 2nd paragraph and page 11,1st
paragraph).
Regarding claim 8 combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition CN82 teaches, wherein the at least one machine learning model predicts the load parameter value of the electrical equipment for the future time based on the plurality of load parameter values and at least one temperature parameter associated with the electrical equipment (during predicting load, in addition to historical load data, transformer oil temperature data is also considered by the machine learning model, page 13th,4th paragraph, page 11, 2nd and 3rd paragraph).
Regarding claim 9 combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition CN82 teaches, further comprising predicting, by the processor circuit (processor, page 16, 3rd paragraph), at least one hot-spot temperature value for a component of the electrical equipment for the future time based on the predicted load parameter value6 (predict maximum oil temperature of the transformer in addition to load forecast using historical load parameters and oil temperature, page 13, 4th paragraph).
Regarding claim 10 combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition JP34 teaches, further comprising predicting, by the processor circuit, at least one overload capacity value for the electrical equipment based on the predicted load parameter value (based on power flow calculation which involves predicted load in future, overload margin for the power system for future time is determined page 10,4th and 5th paragraph), the at least one overload capacity value associated with at least one time period subsequent to the future time (the determined overflow margin is set for the next 30-60 minutes in future that is one time period subsequent to the future time, page 12, 2nd and 3rd paragraph).
Regarding claim 13 combination of JP34, Raith et al. and CN82 teach the method for predicting load parameter, calculating overload and changing at least one parameter. Therefore together they teach the monitoring device performing the functional steps of predicting load parameter, calculating overload and changing at least one parameter of the claimed method as described in claim 1. Claim 13 has additional limitations which are taught by JP34, a processor circuit (processing unit having different unit such as system state unit for prediction which are realized by a processor such as CPU7, page 7, 3rd paragraph); and
a memory comprising machine readable instructions that, when executed
by the processor circuit, cause the processor circuit (program in the form of a file is
in a executable format in a memory or being recorded in a recorded medium that
program instructions stored in a memory or recording medium, page 25, 1st paragraph).
Regarding claims 15 and 16 combination of JP34, Raith et al. and CN82 teach the method for predicting load parameter, calculating overload and changing at least one parameter. Therefore together they teach the monitoring device performing the
functional steps of predicting load parameter, calculating overload and changing at least
one parameter of the claimed method as described in claims 5 and 6.
Regarding claim 17 combination of JP34, Raith et al. and CN82 teach the method for predicting load parameter, calculating overload and changing at least one parameter. Therefore together they teach the processor circuit performing the functional steps of predicting load parameter, calculating overload and changing at least one parameter of the claimed method as described in claim 1. Claim 17 has additional limitations which are taught by CN82, a non-transitory computer readable medium comprising instructions when executed by a processor circuit (computer readable instructions stored in computer memory which can be loaded on to computer or other programmable data processing device (processing circuit) page 16, 4th paragraph and page 17, 1st paragraph).
Regarding claims 19 and 20 combination of JP34, Raith et al. and CN82 teach the method for predicting load parameter, calculating overload and changing at least one parameter. Therefore together they teach the processor circuit performing the functional steps of predicting load parameter, calculating overload and changing at least one parameter of the claimed method as described in claims 5 and 6.
Claim(s) 2,3,14 and 18 rejected under 35 U.S.C.103 as being unpatentable over JP34 (JP 2019216534 A) in view of Raith et al. (US 20200312538 A1) and CN82 (CN 108664682 A) and Cai et al. (US 20210281077 A1).
Regarding claim 2 combination of JP34, Raith et al. and CN82 teach the method of claim 1.
Neither in combination nor individually JP34, Raith et al. and CN82 teach the details of wherein the at least one machine learning model is trained based on a plurality of determined relationships between a predefined number of load parameter values and at least one subsequent load parameter value from a time series data stream obtained for a predetermined period of time.
Cai et al. teaches, wherein the at least one machine learning model is
trained based on a plurality of determined relationships between a predefined
number of load parameter values and at least one subsequent load parameter
value from a time series (load variation over specific time period, [0030]) data stream
obtained for a predetermined period of time (based on predicted load profile which
has predetermined load values over time, demand and amount of energy the equipment
can provide can be predicted for a subsequent time period. The machine learning model
trains the load profile model using plurality of load profiles (determined relationships)
which has load variation over time (time series) as recited in [0078] associated with the power generation system to predict load for subsequent period, [0042],[0043] and
[0079]).
Therefore it would have been obvious before the effective filing date of the
claimed invention to a person of ordinary skill in the art to modify the method for
predicting load parameters values using machine learning model by as taught by
combination of JP34 and CN82 by applying the known technique of training the
machine learning model with load parameters values and subsequent load parameter
as taught by Cai et al. as an improvement to machine learning to yield predictable
results for predicting load and overload conditions using trained machine learning
model.
Regarding claim 3, combination of JP34, Raith et al.,CN82 and Cai et al. teach the method of claim 2. In addition Cai et al. teaches, wherein the plurality of determined relationships is validated based on a comparison8 of at least one expected load parameter value derived from the predefined number of load parameter values and the at least one subsequent load parameter value (the machine learning model training load profile performs cross validations between multiple load profiles and predicted load parameters values and selects the training set (hyperparameter set) with the best overall cross validation score that is the machine learning model can predict accurate load for the subsequent time period based on predetermined load profiles for a power system equipment, [0103], [0104] and [0043]).
Regarding claim 14 combination of JP34, Raith et al.,CN82 and Cai et al. teach the method where machine learning model is trained based on plurality of relationship. Therefore together they teach the device performing the functional steps of training the machine learning model based on plurality of relationship of the claimed method as described in claim 2.
Regarding claim 18 combination of JP34, Raith et al., CN82 and Cai et al. teach the method where machine learning model is trained based on plurality of relationship. Therefore together they teach the device/computer readable medium performing the functional steps of training the machine learning model based on plurality of relationship of the claimed method as described in claim 2.
Claim 4 is rejected under 35 U.S.C.103 as being unpatentable over JP34 (JP 2019216534 A) in view of Raith et al. (US 20200312538 A1) and CN82 (CN 108664682 A) and Vitullo (US 20180113482 A1).
Regarding claim 4 combination of JP34 and CN82 teach the method claim 1. In
addition CN82 teaches, wherein predicting the load parameter value for the future
time is made successively using the predefined number of load parameter values
as an input set to the at least one machine learning model (historical load data and
other additional data are input to the regression analysis model (machine learning
model) to predict future load, page 10, 5th paragraph, page 11, 2nd and 3rd paragraph).
Neither in combination nor individually, JP34 or CN82 teach the details of
wherein the input set is successively generated with a moving window technique from
the time series data stream of load parameter values.
Vitullo teaches, wherein the input set is successively generated with a
moving window technique from the time series data stream of load parameter
values (predictive modelling system which predicts load for equipment uses sliding
data window technique (moving window) [0090], [0158] and [0159]).
Therefore it would have been obvious before the effective filing data of the
claimed invention to a person of ordinary skill in the art to modify the method predicting
load parameter value using machine learning model as taught by combination of JP34, Raith et al. and CN82 by applying the know technique of training the machine learning model with moving window technique (sliding data window) as taught by Vitullo as an improvement to machine learning model to yield predictable results for predicting load with more efficiency.
Claims 11 and 12 are rejected under 35 U.S.C.103 as being unpatentable over JP34 (JP 2019216534 A) in view of Raith et al. (US 20200312538 A1) and CN82 (CN 108664682 A) and Saers et al. (US 20190027932 A1).
Regarding claim 11, combination of JP34, Raith et al. and CN82 teach the method of claim 1. In addition JP34 teaches, wherein the electrical equipment comprises a transformer (the power system includes a transformer, page 5, 4th paragraph and page 12, 2nd paragraph).
Neither in combination nor individually JP34, Raith et al. and CN82 teach the details of operating the transformer based at least in part on the at least one changed parameter. However JP34 explicitly teaches in page 2, 1st paragraph, page 3, 3rd paragraph, that a control command is initiated for the selected equipment in the power system and the control command is executed by the equipment based predicted load and overload condition. Transformer is part of the power system but it not clear whether specifically the transformer is controlled.
On the other hand Saers et al. teaches, the method further comprising:
operating the transformer based at least in part on the at least one changed
parameter (based on the predicted load/ future power level, precool the transformer).
Therefore it would have been obvious before the effective filing date of the
claimed invention to a person of ordinary skill in the art to modify the method for
predicting load for a power system including a transformer and changing at least one
parameter associated with the electrical equipment as taught by combination of JP34, Raith et al. and CN82 by controlling the transformer based on the determined control command as taught by Saers et al. as an improvement to transformer operation since precooling the transformer postpones the overheating of the transfer during high load as taught by Saers et al. in [0031].
Regarding claim 12, combination of JP34, Raith et al., CN82 and Saers et al. teach the method of claim 11. In addition Saers et al. teaches, wherein operating the
transformer includes operating at least one cooling component (cooling equipment, [0053]) of the transformer in response to the predicted load parameter
value to change a temperature of at least one component of the electrical
equipment prior to the future time (based on the predicted future load, the cooling
equipment of the transformer is controlled to precool the transformer, [0027]-[0031] and
[0053]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
(WO 2015042793 A1) teaches a method for determining overload curve of a transformer based on current transformer operational data. The overload curve indicates at what overload capacity (percentage) in y axis and for how long the transformer can operate in x-axis as shown below in Fig.3 of the reference.
PNG
media_image1.png
426
708
media_image1.png
Greyscale
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 ANZUMAN SHARMIN whose telephone number is (571)272-7365. The examiner can normally be reached M and Th 7:00am - 3:00pm and Tue 8:00am-12:00pm.
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, KAMINI SHAH can be reached at (571)272-2279. 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.
/ANZUMAN SHARMIN/Examiner, Art Unit 2115
/KAMINI S SHAH/Supervisory Patent Examiner, Art Unit 2115
1 In order to predict energy fluctuation of the transformer in future, the load on the transformer in future must also be known otherwise how would the model know the change in energy supplied by the transformer if change in load on the transformer is not known.
2 In view of JP34, the load forecasting model will know about the predicted future accident determined based on load on transformer in the future, energy fluctuation on the transmission line and on the transformer in the future time.
3 Power demand is equivalent to load on the transformer or in other words define how much energy the transformer has to provide to meet demand or load in future.
4 Overload capability or overload tolerance is determined based future power demand/load on transformer and energy fluctuation.
5 Determining amount of overload capacity the transformer can be operated in future.
6 There are two ways the temperature value of the equipment can be forecasted, either by using load
parameters only or using a combination of load parameters, temperature parameters and other related
data. Each choice is an obvious variation of each other since there is no specific way recited on the claim
that would exclude to use combination of data instead of just load parameters. Also the claim does not
state how only using load parameters temperature is predicted that would exclude the combination of
other data. It will be obvious to someone of ordinary skill in the art to choose from finite number of identified predictable solutions such as only load parameters data or combination of data with reasonable
expectation of success. MPEP.2143.1.(E).
7 Processor with processing circuit.
8 Only recites to compare which values which are load parameters and subsequent load parameter but
did not recite what are those values compared to for validation.
Prosecution Insights