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 .
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1,2,4,5 and 7-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Chemali et al. (US 20200081070 A1).
As per claims 1, Chemali et al. teach
A system comprising:
one or more energy storage units; (Fig. 1, #150, battery powered system, para 64, Fig. 1, #160, plant); and
a computing device (Fig. 1 #110, controller, Fig. 4 c, control computer with database) comprising at least one processor (Fig. 4 c, control computer with database) and at least one memory (Fig. 4 c, control computer with database, Fig. 3, memory cell,) storing instructions that (estimation system 100 is configured to make a determination of a state of charge SOC of battery 102, para 64, Fig. 1), when executed by the at least one processor ( a controller 110 is coupled to battery 102, and a plant 160, and 110 mediates power transfer between 102 and 162, para 64, Fig. 1) cause the computing device to:
receive operational data associated with an energy storage unit one or more energy storage units for a period of time ;(The SOC estimator 106 receives attributes 104 from the sensors 112 and 113, para 66, Fig. 1. The attributes are temperature, pressure monitored by the sensor 112, and voltage monitored by the sensor 113. The sensors 112 and 113 also provide a time sample digital signal, para 65, Fig. 1);
determine one or more cycles of charging and discharging of the energy storage unit during the period of time (In order to determine the weights of the function N, a training data set has been used that include a collective sequence corresponds to repeated charge -discharge cycles of the battery system, para 62);
generate, based on the operational data, a plurality of load collectives (Quantitative attributes of a battery includes time-sampled digital signal, instance temperature, pressure, voltage, and current passing over the conductor, para 65), wherein each load collective of the plurality of load collectives includes:
one or more criteria associated with operation of the energy storage unit;( The test data for the training validation for the current neural network was set to, temperature:0 -25 degree C, a constant current charge rate 2.9 A(1C), followed by a voltage charge at 4.2 V and, was terminated when the current fell below 50 mA, para 95, Fig. 4);
a quantity of cycles, of the one or more cycles, that satisfy the one or more criteria (During experimentation, the battery was exposed to 10 drive cycles, para 96);
determine one or more operational parameters of the energy storage unit for the period of time (A three hours interval or period was set in between the thermal chamber temperature and the battery internal temperature settings, para 95, Fig. 4 b);
provide, to a machine learning model (LSTM-RNN, long short term recurrent neural network, ARN, artificial neural network, para 69), the one or more operational
parameters (voltage, current, temperature, charge, para 17) and one or more load collectives (battery attributes, para 17) of the plurality of load collectives (battery attributes, SOC, para 25);
generate, based on the machine learning model, a predicted capacity of the energy storage unit at an end of the period of time; and configure, based on the predicted capacity, the one or more energy storage units. (Recurrent neural network-RNN being used by the SOC estimator 106 and includes SOC between 0% to 100% that represents the percentages of full capacity remaining in the charge of the battery, an estimate of internal resistance of the battery and the remaining lifetime of the battery, para 66).
As per claims 15 and 18, they contain similar limitations as claim 1 above. Please refer to the analysis above.
As per claim 2, Chemali et al. teach
The system of claim 1, wherein the operational data includes one or more of:
a temperature of the energy storage unit during each of a plurality of intervals of the period of time (The training and validation data included temperature variation. The temperature was set to the range 0 to 25 degree for the battery drive cycles, a three hours interval or cycles were applied in between the thermal chamber and the battery internal temperature settings to 25-degree C, para 95, Fig. 4b);
a voltage level of the energy storage unit during each of the plurality of intervals of the period of time (The battery was fully charged with constant current charge rate of 2.9 A (1C) followed by a voltage charge at 4.2 V for a three hours interval, para 95, Fig. 4b);
a state of charge of the energy storage unit during each of the plurality of
intervals of the period of time;( SOC state of charge of a battery 102 is being determined by the SOC estimation system 100, para 64 Fig. 1. The SOC estimation system 100 includes sensors 112, 113 to monitor the battery 102 and provide quantitative attributes associated with the battery over time as a time sampled signal, para 65);
an amount of current of the energy storage unit during each of the plurality of
intervals of the period of time. (The second sensor 113 monitors the power connections 112 and provides attributes including voltage and current passing over the conductors over time, para 65).
As per claim 4, Chemali et al. teach
The system of claim 1, wherein the one or more criteria associated with operation
of the energy storage unit comprise one or more of:
a range of a temperature of the energy storage unit (ambient temperatures ranges 0 to 25 degrees, the temperatures were set to 0, 10, and 25-degree C, Fig. 4b, para 95);
a range of a voltage level of the energy storage unit (Min/Max voltage :2.5V/4.2V, para 83);
a range of a minimum state of charge of the energy storage unit (minimum SOC is 30% Fig, 25, para 51, Fig. 28);
a range of a maximum state of charge of the energy storage unit (maximum SOC is 85%, Fig. 25, para 51, Fig. 28);
a range of a sum of amounts of current of the energy storage unit (the second sensor 113 monitors the power connection 112 over the accumulation of current passing over time, para 65, a graph of uncorrupted current validation dataset, Fig. 22);
a range of a sum of current squared of the energy storage unit during a time
segment (to determine state of health SOH estimation performance of the convolution neural network CNN, a mean squared error was computed from all individual errors, para 122);
a range of an amount of current of the energy storage unit (the vector of input include voltage, current, and temperature at time step k, para 65, para 70); or
a range of a depth of charge of the energy storage unit (The charge profile beginning at SOC=30% to ending SOC=85%, para 52, Fig. 25).
As per claim 5, Chemali et al. teach
The system of claim 1, wherein one or more criteria associated with operation of
the energy storage unit for a first load collective of the plurality of load collectives are
different from one or more criteria associated with operation of the energy storage unit
for a second load collective of the plurality of load collectives. (The battery attributes monitored by the first sensor 112 are different than the attributes monitored by the second sensor113, para 65).
As per claim 7, Chemali et al. teach
The system of claim 1, wherein the machine learning model comprises a support
vector machine (during SOC estimation, vector input corresponding to the values of voltage, current, temperature was applied to LSTM-RNN, para 70, using kernel in each layer of CNN, para 119), a relevance vector machine (estimate of SOC measurement for future, para 71, injecting Gaussian noise into the measured battery signal, para 130), or a model based on extreme gradient (using Adam [Kingma and Ba, 2014] optimization method, para 123) boosting.
As per claim 8, Chemali et al. teach
The system of claim 1, wherein the one or more load collectives comprise a subset of the plurality of load collectives (Quantitative attributes of a battery includes time-sampled digital signal, instance temperature, pressure, voltage, and current passing over the conductor, para 65, training of LSTM-RNN is performed on a subset of 10 cycles, para 96).
As per claim 9, Chemali et al. teach
The system of claim 1, wherein the instructions, when executed by the at least
one processor, further cause the computing device to:
determine, based on a feature reduction technique (using LSTM-RNN, long short term recurrent neural network for estimating SOC state of charge, para 69), the one or more load
collectives of the plurality of load collectives (the input data given in the vector component represent voltage, current and temperature, para 69).
As per claim 10, Chemali et al. teach
The system of claim 1, wherein the instructions, when executed by the at least
one processor, further cause the computing device to:
train the machine learning model (CNN training, para 127, Fig. 17) using historical data (dataset 1 of 28 contains 40 reference charge profiles, spanning the entire lifetime of the aged battery cell, para 127) including one or more of:
load collectives of the energy storage unit for a prior period of time before
the period of time (The current, voltage, and temperature sequences of the datasets that is collected from the saved repository, para 127);
operational parameters of the energy storage unit for the prior period of
time (The three signal sequences are concatenated together to form an array of having 3 columns and 256 rows, para 127, Fig. 14, Fig. 17); or
a measured capacity of the energy storage unit at an end of the prior
period of time (Resulting capacity was measured by summing up the multiplied value of current and sampling period, para 129).
As per claim 11, Chemali et al. teach
The system of claim 1, wherein the instructions, when executed by the at least
one processor (Fig. 4 c, control computer with database), further cause the computing device to:
configure (System 100 may be configured to provide an estimate of the SOC of the battery, para 64), based on the predicted capacity (The SOC estimator 106 includes a state of charge SOC number ranging 0% to 100% representing the percentage of full capacity remaining, para 66), the one or more energy storage units
by one or more of: adjusting a pattern (LSTM-RNN architecture where input data voltage, current and temperature are provided into the input data step (vector function) and the SOC estimation are given at every time step k, para 69) for the one or more energy storage units (battery 102, plant 162, Fig. 1) to
dispatch electricity (the controller mediates power transfer between the battery 102 and the plant 162, para 64), or augmenting a capacity of the one or more energy storage units (a data augmentation technique is used by injecting Gaussian noise, para 130, RNN provides the output of the neural network by augmenting the input from the neural network, para 67).
As per claim 12, Chemali et al. teach
The system of claim 11, wherein the pattern includes a plurality of time intervals (Fig. 7, times in seconds)
during which the one or more energy storage units are configured to charge or
discharge at a particular rate (a diagram time series vs SOC, where time series spanning multiple charging and discharging cycle, Fig. 7, para 103).
As per claim 13, Chemali et al. teach
The system of claim 1, wherein the instructions, when executed by the at least
one processor, further cause the computing device to:
calculate a degree of influence (calculating Weights, para 73) of each input item of a plurality of input items to
the machine learning model on an output of the machine learning model (LSTRM-RNN applied toward SOC estimation and the train data involved input vector with components voltage, currents, and temperature including input, forget, and output gates and memory cell, para 70, para 73).
As per claim 14, Chemali et al. teach
The system of claim 1, wherein the instructions, when executed by the at least
one processor, further cause the computing device to:
determine a degree of similarity (using a data augmentation technique/Gaussian Noise to battery measured signals, para 130)
between the energy storage unit and another
energy storage unit; and
based on determining that the degree of similarity satisfies a threshold (an offset of up to plus- minus 150 mA and gain of up to plus -minus 3% applied to the current measurements, and similar techniques applied to voltage measurements with plus-minus 5 mV and with plus -minus 5-degree C to temperature, para 130) use
training data for the machine learning model to train a machine learning model for the
other energy storage unit (Alternate copies of training data area created with varying levels of noise and offsets and gains, para 130).
As per claim 16, Chemali et al. teach
The method of claim 15, wherein the machine learning model comprises a
vector machine (during SOC estimation vector input corresponding the values voltage, current, and temperature was applied to LSTM-RNN, para 70, using kernel in each layer of CNN, para 119), a relevance vector machine (estimate of SOC measurement for future, para 71), or a model based on extreme gradient (using Adam [Kingma and Ba, 2014] optimization method, para 123) boosting.
As per claim 17, Chemali et al. teach
The method of claim 15, further comprising:
determining, based on a feature reduction technique (using LSTM-RNN, long short term recurrent neural network for estimating SOC State of charge, para 69), the one or more load
collectives of the plurality of load collectives (the input data given in the vector component represent voltage, current and temperature, para 69).
As per claim 19, Chemali et al. teach
The non-transitory computer-readable medium (A computer, para 153, control computer with database, Fig. 4 c, Fig. 1, #110) of claim 18, wherein the machine
learning model comprises a support vector machine, a relevance vector machine, or a
model based on extreme gradient boosting.
Please see analysis for claim 16 above.
As per claim 20, it contains similar limitations as claim 17 above. Please refer to that analysis.
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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or
nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 3 and 6 are rejected under 35 U.S.C. 103 as being unpatentable over Chemali et al. (US 20200081070 A1) as applied to claims above, and further in view of Nygren et al. (US 20190378354 A1).
As per claim 3, Chemali et al. teach
The system of claim 1, wherein the instructions, when executed by the at least
one processor, further cause the computing device to:
determine the one or more cycles of charging and discharging of the energy
storage unit during the period of time (During an experimentation the battery was exposed to 10 drive cycles and charging test experiment was done in two steps with the first one at 25 degree and the second one from 10 to 25 degree used to validate long short-term memory recurrent neural network. The power profile for the drive cycles has discharge power as great as 40 W per cell and charge power as 35 W per cell, para96. Adam [Kingma and Ba, 2014] method was used for single cycle forward pass and backward pass, para 123).
Chemali et al. teach a cycle counting method for charging and discharging cycles using Adam [ Kingma Ba, 2014] method where they apply single cycle forward pass and backward pass. However, they do not apply the rain-flow counting method in their algorithm.
However, Nygren et al. in the same field of endeavor teach an algorithm using rain-flow counting method in a data storage monitoring system that uses computer, processor and memory system. Nygren et al. use rain-flow counting method for counting pressure cycles on an accumulator based on pre-charge condition and re-charge condition. (Para 307, para 308, Nygren et al.). They apply rain flow counting algorithm to the data record and generate the number of pressure cycles. (para 316, Nygren et al.).
It would have been obvious to a person ordinary skilled in the art, before the effective filing date of the claimed invention, to combine the teaching of Nygren et al. counting pressure cycles, and to include into the counting, charging and discharging cycles taught by Chemali et al. This would have been obvious because both Nygren et al. and Chemali et al. use storage system that include a computing device, processor, and memory and both use a method for cycle counting. By adding the counting method for pressure cycle taught by Nygren et al. into the charging and discharging cycle counting method taught by Chemali et al., the estimation for the state of charge will be more accurate because the rain flow counting algorithm taught by Nygren et al. includes using a calendar time CT. When CT encounters the threshold, it generates a status indicator that the accumulator is due for re-charging pressure (para 307, Nygren et al.), which would provide better accuracy.
As per claim 6, Chemali et al. teach
The system of claim 1, wherein the one or more operational parameters of the
energy storage unit comprise one or more of:
a quantity of cycles, of charging and discharging of the energy storage unit
during the period of time (During experimentations, the battery exposed to 10 drive cycles. In order to validate long short term recurrent neural network, charging test cases was set and also discharge power as set to the power profile, para 96, Fig. 5, para 106, Fig 9);
a quantity of equivalent full cycles of charging and discharging of the energy
storage unit during the period of time (the network LSTM-RNN is trained up to 8 mixed cycles with 2 discharge test case and one charging test cases, para 101);
a sum of amounts of current of the energy storage unit during the period of time (the computation time to train LSTM-RNN network is about 4 hours, para 101);
an average state of charge of the energy storage unit during the period of time (The SOC is estimated to have mean error less than 4%, para 144, Fig. 26);
a length of the period of time (the estimation took total of 4 hours, para 101); or
a capacity of the energy storage unit at a beginning of the period of time (a graph of recorded capacity-based SOH, para 42, Fig. 16).
Although Chemali et al. teach most the limitations in claim 6, they do not specify using the rain flow counting algorithm.
Please see analysis for claim 3 above.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Rokeya Alam whose telephone number is (571)270-0083. The examiner can normally be reached on 7:30am - 4:30pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Mr. Scott Baderman can be reached at telephone number (571-272-3644). The fax phone number for the organization where this application or proceeding is assigned is (571) 273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/ROKEYA SHAWALI ALAM/Examiner, Art Unit 2118
/SCOTT T BADERMAN/Supervisory Patent Examiner, Art Unit 2118