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 .
Claim Rejections - 35 USC § 112
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.
Claims 6 and 14-15 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 6 recites the limitation "the cluster" in line 2. There is insufficient antecedent basis for this limitation in the claim. Claim 5 from which claim 6 depends discloses “a plurality of clusters,” “k-means clustering,” and “a preset cluster.” For the purposes of the present examination, “the preset cluster” is construed. However, further clarification is required.
Claim 14 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite. Claim 14, lines 1-4 disclose “wherein the determining of the second data comprises calculating a plurality of clusters through k-means clustering on the principal component data and determining the differential data as the second data when the differential data is included in a preset cluster.” It is unclear if “when the differential data is included in a preset cluster” is attributed only to the “determining the differential data as the second data” or also includes the “determining of the second data comprises calculating a plurality of clusters through k-means clustering on the principal component data.” For the purposes of the present examination, both are construed as occurring only when the differential data is included in a preset cluster. However, further clarification is required.
Claim 15 recites the limitation "the cluster" in line 2. There is insufficient antecedent basis for this limitation in the claim. Claim 14, from which claim 15 depends, discloses “a plurality of clusters,” “k-means clustering,” and “a preset cluster.” For the purposes of the present examination, “the preset cluster” is construed. However, further clarification is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
The claims are evaluated for patent subject matter eligibility under 35 U.S.C. 101 using the 2019 Revised Patent Subject Matter Eligibility Guidance (2019 PEG) as follows:
Step 1:
Claims 1-10 are directed to an apparatus and therefore falls within the four statutory categories of subject matter.
Step 2A:
This step asks if the claim is directed to a law of nature, a natural phenomenon (product of nature) or an abstract idea. Step 2A is a two-prong inquiry: in prong 1 it is determined whether a claim recites a judicial exception, and if so, then in prong 2 it is determined if the recited judicial exception is integrated into a practical application of that exception.
Analyzing claim 1 under prong 1 of step 2A, the language:
A battery capacity estimation apparatus comprising:
determining voltage data as first data when a logging pattern of the voltage data deviates from a preset reference range;
determining a second data through statistical analysis on the voltage data; and
estimating a capacity by applying data to capacity estimation models generated separately for the first data and the second data.
has a scope that encompasses mental steps, e.g., concepts that may be performed in the human mind; e.g., human observation/performable with pen and paper/mere data gathering. Claim 1 discloses A battery capacity estimation apparatus comprising; construed as a preamble setting forth intended use; determining voltage data as first data when a logging pattern of the voltage data deviates from a preset reference range; construed as a mental step; e.g., observation; determining a second data through statistical analysis on the voltage data; and; construed as a mental step; e.g., performable with pen and paper; estimating a capacity by applying data to capacity estimation models generated separately for the first data and the second data; construed by the examiner as a mental step; e.g. performable with pen and paper. The broadest reasonable interpretation of the abovementioned steps in light of the specification has a scope that encompasses steps that may be performed in the human mind. It is therefore concluded under prong 1 of step 2A that claim 1 recites a judicial exception in the form of an abstract idea, i.e., mental steps. See MPEP 2106.04(a)(2)(A-C) and MPEP 2106.05(f).
In prong 2 of step 2A it is determined whether the recited judicial exception is integrated into a practical application of that exception by: (1) identifying whether there are any additional elements recited in the claim beyond judicial exception(s); and (2) evaluating those additional elements individually and in combination to determine whether they integrate the exception into a practical application.
Analyzing claim 1 under prong 2 of step 2A, in addition to the abstract ideas described above, claim 1 further recites:
a voltage measuring unit measuring a voltage of a battery cell
of the battery cell
of the battery cell
of the battery cell
for the battery cell that is a measurement target
a filtering unit
a statistical analyzing unit
a capacity estimating unit
classified by the filtering unit or the statistical analyzing unit
Regarding additional elements 1-5, although these elements appear to represent physical structures, mere physicality or tangibility is not a relevant consideration in Step 2A, prong 2. See MPEP 2106.04(d).I. Although the way in which the additional elements use or interact with the exception may integrate it into a practical application (see, MPEP 2106(d).III.), claim 1 does not appear to recite any meaningful interaction between elements 1-5 and the abstract ideas applied by the voltage measuring unit measuring a voltage of a battery cell that is a measurement target which merely providing data necessary to implement the abstract ideas. Although additional elements 1-5 are recited with sufficient specificity such that they might be regarded as a particular machine, the machine does not implement the abstract idea and its involvement appears to be only extra-solution activity in the form of mere data gathering. Use of a machine that contributes only nominally or insignificantly to the execution of the claimed method (e.g., in a data gathering step or in a field-of-use limitation) does not integrate a judicial exception or provide significantly more. See, e.g., MPEP 2106.05(b)(III).
The additional elements of 6-9 discussed above in connection with prong 2 of step 2A merely represents implementation of the abstract idea using a generic processor/computer and use of a generic processor/computer. However, use of a computer or other machine in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general-purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more. See MPEP 2106.05(f)
For these reasons, additional elements 1-9, considered alone or in combination with the abstract ideas, do not appear sufficient to integrate the abstract ideas into a practical application under step 2A, prong 2.
Step 2B:
Considered individually and in combination, additional elements 1-9 do not appear to provide significantly more than the abstract ideas for reasons analogous to those discussed above at step 2A, prong 2. Additionally, with respect to additional elements 1-5, utilizing a battery capacity estimation apparatus comprising a voltage measuring unit measuring a voltage of a battery cell that is a measuring target was well-known, routine, and conventional before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. See, e.g., US 2016/0195587 A1 to Lee, particularly to figs. 1-2 and paras. [0050]-[0054]; See also US 11,422,191 B2, particularly to figs. 1-2 and 6, as well as col. 5, lines 35-67 and col. 11, line 22 – col. 12, line 10; See also US 8,519,674 B2 to fig. 1 and col. 2, lines 14-33 and col. 2, line 65 – col. 3, line 54. Accordingly, additional elements 1-5 appear to represent no more than a well-understood, routine, conventional data-gathering activity.
For these reasons, additional elements 1-9, considered alone or in combination with the abstract ideas, do not appear to provide significantly more than the abstract ideas under step 2B. Dependent claims 2-10 appear to relate exclusively to aspects of the abstract ideas of claim 1 and therefore do not represent integration of the abstract ideas into a practical application or represent significantly more than the abstract ideas themselves.
Step 1:
Claims 11-15 are directed to a method and therefore falls within the four statutory categories of subject matter.
Step 2A:
This step asks if the claim is directed to a law of nature, a natural phenomenon (product of nature) or an abstract idea. Step 2A is a two-prong inquiry: in prong 1 it is determined whether a claim recites a judicial exception, and if so, then in prong 2 it is determined if the recited judicial exception is integrated into a practical application of that exception.
Analyzing claim 11 under prong 1 of step 2A, the language:
A battery capacity estimation method comprising:
determining voltage data as first data when a logging pattern of the voltage data deviates from a preset reference range;
determining second data through statistical analysis on the voltage data; and
estimating a capacity by applying the first data and the second data to capacity estimation models, respectively.
has a scope that encompasses mental steps, e.g., concepts that may be performed in the human mind; e.g., human observation/performable with pen and paper/mere data gathering. Claim 11 discloses A battery capacity estimation method comprising; construed as a preamble setting forth intended use; determining voltage data as first data when a logging pattern of the voltage data deviates from a preset reference range; construed as a mental step; e.g., observation; determining second data through statistical analysis on the voltage data; and; construed by the examiner as a mental step; e.g., performable with pen and paper; estimating a capacity by applying the first data and the second data to capacity estimation models, respectively; construed by the examiner as a mental step; e.g., performable with pen and paper. The broadest reasonable interpretation of the abovementioned steps in light of the specification has a scope that encompasses steps that may be performed in the human mind. It is therefore concluded under prong 1 of step 2A that claim 11 recites a judicial exception in the form of an abstract idea, i.e., mental steps. See MPEP 2106.04(a)(2)(A-C) and MPEP 2106.05(f).
In prong 2 of step 2A it is determined whether the recited judicial exception is integrated into a practical application of that exception by: (1) identifying whether there are any additional elements recited in the claim beyond judicial exception(s); and (2) evaluating those additional elements individually and in combination to determine whether they integrate the exception into a practical application.
Analyzing claim 11 under prong 2 of step 2A, in addition to the abstract ideas described above, claim 11 further recites:
measuring a voltage of a battery cell
of the battery cell
of the battery cell
of the battery cell
Regarding additional elements 10-13, although these elements appear to represent physical structures, mere physicality or tangibility is not a relevant consideration in Step 2A, prong 2. See MPEP 2106.04(d).I. Although the way in which the additional elements use or interact with the exception may integrate it into a practical application (see, MPEP 2106(d).III.), claim 11 does not appear to recite any meaningful interaction between elements 10-13 and the abstract ideas applied by the battery capacity estimation method comprising measuring a voltage of a battery cell which is merely providing data necessary to implement the abstract ideas. Although additional elements 10-13 are recited with sufficient specificity such that they might be regarded as a particular machine, the machine does not implement the abstract idea and its involvement appears to be only extra-solution activity in the form of mere data gathering. Use of a machine that contributes only nominally or insignificantly to the execution of the claimed method (e.g., in a data gathering step or in a field-of-use limitation) does not integrate a judicial exception or provide significantly more. See, e.g., MPEP 2106.05(b)(III).
For these reasons, additional elements 10-13, considered alone or in combination with the abstract ideas, do not appear sufficient to integrate the abstract ideas into a practical application under step 2A, prong 2.
Step 2B:
Considered individually and in combination, additional elements 10-13 do not appear to provide significantly more than the abstract ideas for reasons analogous to those discussed above at step 2A, prong 2. Additionally, with respect to additional elements 10-13, a battery capacity estimation method comprising measuring a voltage of a battery cell was well-known, routine, and conventional before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. See, e.g., US 2016/0195587 A1 to Lee, particularly to figs. 1-2 and paras. [0050]-[0054]; See also US 11,422,191 B2, particularly to figs. 1-2 and 6, as well as col. 5, lines 35-67 and col. 11, line 22 – col. 12, line 10; See also US 8,519,674 B2 to fig. 1 and col. 2, lines 14-33 and col. 2, line 65 – col. 3, line 54. Accordingly, additional elements 10-13 appear to represent no more than a well-understood, routine, conventional data-gathering activity.
For these reasons, additional elements 10-13, considered alone or in combination with the abstract ideas, do not appear to provide significantly more than the abstract ideas under step 2B. Dependent claims 12-15 appear to relate exclusively to aspects of the abstract ideas of claim 11 and therefore do not represent integration of the abstract ideas into a practical application or represent significantly more than the abstract ideas themselves.
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, 7, and 9-11 are rejected under 35 U.S.C. 103 as being unpatentable over C. Liu, J. Tan, H. Shi and X. Wang, "Lithium-Ion Cell Screening With Convolutional Neural Networks Based on Two-Step Time-Series Clustering and Hybrid Resampling for Imbalanced Data," in IEEE Access, vol. 6, pp. 59001-59014, 2018, doi: 10.1109/ACCESS.2018.2875514, hereinafter Liu, in view of Lee (US 2016/0195587 A1), hereinafter Lee
Regarding claim 1, Liu discloses A battery capacity estimation apparatus comprising:
a data source providing a voltage of a battery cell; (Liu, e.g., see pg. 59003, to fig. 1 illustrating a formation and screening process which depicts a 1st capacity screening and a 2nd discharge voltage curve screening; see also pg. 59002, col. 2 disclosing a two-stage cell screening method, namely, capacity screening and discharge screening, is implemented in an orderly manner. First, we initially divide the cells into six classes according to their capacity range. Then, the cells in each class are used by TTSCHR-CNN [two-step time-series clustering and hybrid resampling – convolutional neural network] to screen for defective cells in the discharge voltage curves. Generally, the discharge voltage curve of a cell can be considered as a time series. Due to the higher time and space complexity of unsupervised time-series clustering in a large number of samples, we first consider labeling the discharge voltage time-series dataset of cells as the majority class with the label “1” (negative samples) and the minority class with the label “0” (positive samples, the discharge voltage time series of defective cells) offline by using a two-step time-series clustering method and then use a hybrid resampling method to obtain a data set with balanced positive and negative samples to train the convolutional neural network (CNN) model for online cell screening; see also fig. 7 illustrating a data layer of a plurality of data sources 1-n).
a filtering unit determining voltage data as first data when a logging pattern of the voltage data of the battery cell deviates from a preset reference range; (Liu, e.g., see rejection as applied above, specifically to fig. 1 illustrating 2nd discharge voltage curve screening, wherein the illustration depicts normal and abnormal curves of voltage over time; see also pg. 59003, col. 1-2 disclosing in the second DVC [discharge voltage curve] screening process, the cells in each class (class 3 in fig. 1) continue to be screened by evaluating the similarity of the discharge voltage curves, and defective cells with large deviations in their discharge voltage curves, the 5 red dotted lines in Figure 1 (their capacity distribution is indicated by the red circles), are determined to be the abnormal DVCs. Generally, the DVC of a cell can be considered a time series, which is a sequence of real-value data points with timestamps. Therefore, DVC screening is a time-series classification problem. In this paper, we denote a discharge voltage time series as
V
=
v
1
,
v
2
,
…
,
v
n
where
v
i
is the discharge voltage value at time stamp
t
i
, and there are
n
timestamps for each discharge voltage time series. We denote a labeled discharge voltage time series data as
D
=
{
V
i
,
y
i
}
i
=
1
N
, which contains
N
discharge voltage time series and their associated labels. For each
i
=
1,2
,
…
,
N
,
V
i
represents the
i
t
h
discharge voltage time series, and its label is
y
i
. Herein ,
y
i
is a class label in
C
=
1,0
, where “1” represents the normal discharge voltage time series and “0” represents the abnormal discharge voltage time series; see also fig. 2 illustrating the Raw DVCs of fig. 1 which are then separated out to DVCs with “1” and DVCs with “0”; examiner notes the model demonstrating DVCs with nomenclature “0” of fig. 2 is construed as the filtering unit; see also figs. 4 and 7).
a statistical analyzing unit determining second data through statistical analysis on the voltage data of the battery cell; and (Liu, e.g., see rejection as applied above with regard to figs. 1-2 and 4, specifically to pg. 59002, col. 2 disclosing two-step time-series clustering and hybrid resampling for imbalanced data (TTSCHR-CNN); examiner notes that time series clustering is necessarily a statistical analysis; see also pg. 59003, col. 2 cited above disclosing herein ,
y
i
is a class label in
C
=
1,0
, where “1” represents the normal discharge voltage time series and “0” represents the abnormal discharge voltage time series; see also fig. 2 illustrating the Raw DVCs of fig. 1 which are then separated out to DVCs with “1” and DVCs with “0”; examiner notes DVCs with “1” of fig. 2 is construed as second data; see also pg. 59004, col. 2 – pg. 59005, col. 1 disclosing the overall architecture of TTSCHR-CNN for DVC screening is depicted in Figure 2 and consists of two stages: offline and online. In the offline stage, we can classify the raw DVCs into two classes by using a two-step time-series clustering method: the normal DVCs with the label “1” and the abnormal DVCs with the label “0”; examiner notes the model demonstrating DVCs with nomenclature “1” of fig. 2 is construed as the statistical analyzing unit).
a capacity estimating unit estimating a capacity of the battery cell by applying data, classified by the filtering unit or the statistical analyzing unit for the battery cell that is a measurement target, to capacity estimation models generated separately for the first data and the second data. (Liu, e.g., see rejection as applied above with respect to figs. 1-2 and 4; see also pg. 59006, col. 2 – pg. 59007, col. 1 disclosing resampling techniques can be divided into three groups; oversampling methods, undersampling methods, and hybrid methods. Third, hybrid methods are a combination of oversampling and undersampling methods. In this paper, a hybrid resampling method, which creates some synthetic discharge voltage time-series samples with the label “0” by using SMOTE [synthetic minority over-sampling technique] and discards a number of the similar samples with the label “1” by using undersampling based on clustering, is proposed to obtain a balanced training set for the CNN model; wherein the CNN is construed as the capacity estimating unit; see also fig. 5 illustrating five classes of cluster center curves and discharge voltage curves (DVCs) of the minority class divided by a two-step time-series clustering; see also tables 2-3;see also pg. 59008, col. 2 – pg. 59010, col. 2 disclosing after data alignment, five classes of discharge voltage time series of the cells with capacities between 1.98 Ah and 2.18 Ah are labeled as the majority class samples with the label “1” or the minority class samples with the label “0” by the two-step time-series clustering (TTSC) method in section III/C, where
k
in the k-means algorithm is set to 10. Specifically, each class of discharge voltage time series in the five classes is classified into 10 clusters by using the k-means algorithm, and five dividing lines are defined by the two most adjacent cluster centers of each class, as shown in fig. 5(a.x). Ten clusters in each class are divided into two parts: minority clusters containing few samples and majority clusters containing most of the samples. For example, in Figure 5(a1), seven dotted lines (the number of cluster centers is 1, 4, 5, 6, 8, 9, and 10) are defined as the minority clusters, three solid lines (the number of cluster centers is 2, 3, and 7) are defined as the majority clusters, and No. 3 and No. 9 as the two most adjacent cluster centers are selected to define the dividing line. The samples below the dividing lines and the abnormal samples manually screened above the dividing lines are defined as the minority class samples, and the DVCs of the minority class are shown in Figure 5(b.x). Hence, the new datasets with balanced positive and negative samples are used to train the CNN models for online lithium cell screening, where the datasets are divided into training sets for learning the CNN models and test sets for evaluating the CNN models. The numbers of samples in different data preprocessing steps are shown in Table 2. The negative and positive samples of training sets for the five CNN models in Table 2 are basically balanced. The five test sets contain 1000 negative samples and 100 positive samples. We train the CNN models using TensorFlow, and the performance metrics of five CNN models on the test sets are summarized in Table 3; see also fig. 6 illustrating the discharge voltage curves (DVCs) of abnormal cells predicted by the trained CNN models on the unlabeled test sets. (a1) DVCs of unlabeled test set with capacities between 1.98 and 2.02 Ah. (a2) DVCs of unlabeled test set with capacities between 2.02 and 2.06 Ah. (a3) DVCs of unlabeled test set with capacities between 2.06 and 2.10 Ah. (a4) DVCs of unlabeled test set with capacities between 2.10 and 2.14 Ah. (a5) DVCs of unlabeled test set with capacities between 2.14 and 2.18 Ah; see also Table 4; see also fig. 7 illustrating the application architecture of online lithium-ion cell screening in industrial production).
Liu may be relied upon as disclosing capacity screening and discharge voltage curve screening, which is acquired from a data source, and therefore Liu may not be relied upon as explicitly disclosing a voltage measuring unit measuring a voltage.
However, Lee further discloses voltage measuring unit measuring a voltage. (Lee, e.g., see fig. 1 illustrating battery cells (110) and battery information measurer (120); see also paras. [0050]-[0051] disclosing a battery management system (BMS) senses data of any or any combination of a voltage, current, and a temperature measured form the battery, as only non-limiting examples, and manages the example voltage, current, and/or the temperature. In an example, the BMS may include, or be represented by, the battery pack (110) and the battery information measurer (120). In another example, the BMS may include, or be represented by, the battery pack (110), the battery information measurer (120), and the apparatus (130); see also para. [0056] disclosing the battery information measurer (120) measures battery information from each, for example, of the battery cells in the battery pack (110)).
Accordingly, it would be prima facie obvious to one of ordinary skill in the art, at the time the invention was effectively filed, to have modified Liu with Lee’s voltage measuring unit measuring a voltage for at least the reasons that a voltage measurement of a particular battery or battery packs may be performed, as taught by Liu; e.g., see para. [0052].
Regarding claim 2, Liu in view of Lee discloses The battery capacity estimation apparatus of claim 1, wherein the statistical analyzing unit performs statistical analysis on data other than data determined as the first data by the filtering unit. (Liu, e.g., see rejection as applied to claim 1, specifically to pg. 59006, col. 2 – pg. 59007, col. 1 disclosing hybrid methods are a combination of oversampling and undersampling methods. In this paper, a hybrid resampling method, which creates some synthetic discharge voltage time-series samples; construed as data other than data determined as the first data by the filtering unit; with the label “0” by using SMOTE and discards a number of the similar samples with label “1” by using undersampling based on clustering, is proposed to obtain a balanced training set for the CNN model. Given the discharge voltage time-series samples with the label “0” as the positive learning dataset (the minority class) and those with the label “1” as the negative learning dataset (the majority class),
P
=
V
11
,
V
12
,
…
,
V
1
P
and
N
=
{
V
01
,
V
02
,
…
,
V
0
N
}
, where
N
≫
P
,
V
i
j
∈
R
n
×
1
and
n
denotes the time-series length or dimension. for the minority class dataset
P
, we use an oversampling algorithm based on SMOTE to create new synthetic discharge voltage time-series samples. specifically, for each minority class sample
V
0
j
∈
P
, we use the k-nearest neighbors algorithm to calculate the Euclidean distance between sample
V
0
j
to all samples in P to obtain
k
p
nearest samples, and for each sample
V
^
0
j
in
n
p
samples, we create a new synthetic sample according to the following formula:
V
0
n
e
w
=
V
0
j
+
∆
×
V
^
0
j
-
V
0
j
(4); examiner notes that a synthetic sample calculated from the clustering method disclosed and cited above is construed as other data than the data determined as the first data.
Regarding claim 7, Liu, of Liu in view of Lee, in the current embodiment, is not relied upon as explicitly disclosing: The battery capacity estimation apparatus of claim 1, wherein the capacity estimation model is a long short-term memory networks (LSTM) model.
However, an alternative embodiment of Liu discloses: wherein the capacity estimation model is a long short-term memory networks (LSTM) model. (Liu, e.g., see pg. 59007, cols. 1-2 disclosing LSTM fully convolutional network (LSTM-FCN), and multi-scale convolutional neural network (MCNN) take advantage of CNN to address univariate time series, and a multichannel CNN has been proposed to solve multivariate time series. In particular, LSTM-FCN improved the performance of FCN by augmenting the FCN module and has high performance for time-series classification).
Accordingly, it would be prima facie obvious to one of ordinary skill in the art, at the time the invention was effectively filed, to have modified Liu in view of Lee with an alternative embodiment of Liu for at least the reasons that it is known in the art to utilize LSTM for learning long-term temporal patterns from sequential data, such as the time-dependent degradation of batteries.
Regarding claim 9, Liu in view of Lee discloses The battery capacity estimation apparatus of claim 1, wherein when at least one of a logging time and a number of voltage data of the battery cell deviates from the reference range, the filtering unit determines the voltage data as the first data. (see rejection as applied to claim 1, specifically figs. 1-2 and 4 illustrating a plurality of discrete measurements of voltage over time, forming a curve; see also Liu, e.g., see pg. 59003, col. 2 disclosing generally, the DVC of a cell can be considered a time series, which is a sequence of real-value data points with timestamps; construed as a logging time. Therefore, DVC screening is a time-series classification problem).
Regarding claim 10, Liu in view of Lee discloses The battery capacity estimation apparatus of claim 1, wherein the first data is data in which the voltage data of the battery cell is in a discontinuous form, and the second data is data in which the voltage data of the battery cell is in a continuous form. see rejection as applied to claim 1, specifically pg. 590003, col. 2; see also Liu, e.g., see pg. 59004, col. 1 disclosing in industrial processes, the discharge voltage time series with the label “0” are the minority compared with those with the label “1”. Training classification algorithms from such imbalanced data is a formidable challenge and may lead to a minority class of samples being misjudged as majority classes, thereby reducing the classification accuracy; see also pg. 59006, col. 2 – pg. 59007, col. 1 disclosing typically, the discharge voltage time series with the label “0” represent defective cells and are always a minority class. Training classification algorithms for such an imbalanced dataset may lead to a small number of samples in minority classes being misclassified as majority classes, thereby reducing the classification accuracy. Consequently, resampling techniques, which are used to rebalance the sample space for an imbalanced dataset to alleviate the effect of the skewed class distribution in the learning process, are often performed to address imbalanced learning to obtain a balanced training set. Resampling techniques can be divided into three groups: oversampling methods, undersampling methods, and hybrid methods. First, oversampling methods are used to eliminate the risk of an imbalanced distribution by creating synthetic minority samples. The synthetic minority oversampling technique (SMOTE), which creates artificial data based on the feature space similarities between existing minority samples is a powerful method that has shown great success in various fields. Second, undersampling methods are used to eliminate the risk of an imbalanced distribution by discarding similar samples in the majority class. third, hybrid methods are a combination of oversampling and undersampl9ing methods. In this paper, a hybrid resampling method, which creates some synthetic discharge voltage time-series samples with the label “0” by using SMOTE and discards a number of the similar samples with the label “1” by using undersampling based on clustering, is proposed to obtain a balanced training set for the CNN model; examiner notes that the first data based on the voltage data is necessarily in a “discontinuous form” as it is data not easy to analyze, requiring synthetic samples, while the second data of the voltage data is in “continuous form being less intense to analyze).
Regarding claim 11, Liu discloses A battery capacity estimation method comprising:
providing a voltage of a battery cell; (Liu, e.g., see pg. 59003, to fig. 1 illustrating a formation and screening process which depicts a 1st capacity screening and a 2nd discharge voltage curve screening; see also pg. 59002, col. 2 disclosing a two-stage cell screening method, namely, capacity screening and discharge screening, is implemented in an orderly manner. First, we initially divide the cells into six classes according to their capacity range. Then, the cells in each class are used by TTSCHR-CNN [two-step time-series clustering and hybrid resampling – convolutional neural network] to screen for defective cells in the discharge voltage curves. Generally, the discharge voltage curve of a cell can be considered as a time series. Due to the higher time and space complexity of unsupervised time-series clustering in a large number of samples, we first consider labeling the discharge voltage time-series dataset of cells as the majority class with the label “1” (negative samples) and the minority class with the label “0” (positive samples, the discharge voltage time series of defective cells) offline by using a two-step time-series clustering method and then use a hybrid resampling method to obtain a data set with balanced positive and negative samples to train the convolutional neural network (CNN) model for online cell screening).
determining voltage data as first data when a logging pattern of the voltage data of the battery cell deviates from a preset reference range; The broadest reasonable interpretation of a method (or process) claim having contingent limitations requires only those steps that must be performed and does not include steps that are not required to be performed because the condition(s) precedent are not met; see, e.g., MPEP 2111.04(III); because the step of determining voltage data as first data is only performed if a condition precedent is met; e.g., when a logging pattern of the voltage data of the battery cell deviates from a preset reference range, the broadest reasonable interpretation of this claim does not require this step; e.g., determining voltage data as first data to be performed; accordingly, this step does not carry patentable weight.
determining second data through statistical analysis on the voltage data of the battery cell; and (Liu, e.g., see rejection as applied above with regard to figs. 1-2 and 4, specifically to pg. 59002, col. 2 disclosing two-step time-series clustering and hybrid resampling for imbalanced data (TTSCHR-CNN); examiner notes that time series clustering is necessarily a statistical analysis; see also pg. 59003, col. 2 cited above disclosing herein ,
y
i
is a class label in
C
=
1,0
, where “1” represents the normal discharge voltage time series and “0” represents the abnormal discharge voltage time series; see also fig. 2 illustrating the Raw DVCs of fig. 1 which are then separated out to DVCs with “1” and DVCs with “0”; examiner notes DVCs with “1” of fig. 2 is construed as second data; see also pg. 59004, col. 2 – pg. 59005, col. 1 disclosing the overall architecture of TTSCHR-CNN for DVC screening is depicted in Figure 2 and consists of two stages: offline and online. In the offline stage, we can classify the raw DVCs into two classes by using a two-step time-series clustering method: the normal DVCs with the label “1” and the abnormal DVCs with the label “0”; examiner notes the model demonstrating DVCs with nomenclature “1” of fig. 2 is construed as the statistical analyzing unit).
estimating a capacity of the battery cell by applying the first data and the second data to capacity estimation models, respectively. The broadest reasonable interpretation of a method (or process) claim having contingent limitations requires only those steps that must be performed and does not include steps that are not required to be performed because the condition(s) precedent are not met; see, e.g., MPEP 2111.04(III); because the step of determining voltage data as first data is only performed if a condition precedent is met; e.g., when a logging pattern of the voltage data of the battery cell deviates from a preset reference range, the broadest reasonable interpretation of this claim does not require this step; e.g., estimating a capacity of the battery cell by applying the first data to be performed; accordingly, this step does not carry patentable weight.
Liu may be relied upon as disclosing capacity screening and discharge voltage curve screening, which is acquired from a data source, and therefore Liu may not be relied upon as explicitly disclosing a measuring a voltage.
However, Lee further discloses measuring a voltage. (Lee, e.g., see fig. 1 illustrating battery cells (110) and battery information measurer (120); see also paras. [0050]-[0051] disclosing a battery management system (BMS) senses data of any or any combination of a voltage, current, and a temperature measured form the battery, as only non-limiting examples, and manages the example voltage, current, and/or the temperature. In an example, the BMS may include, or be represented by, the battery pack (110) and the battery information measurer (120). In another example, the BMS may include, or be represented by, the battery pack (110), the battery information measurer (120), and the apparatus (130); see also para. [0056] disclosing the battery information measurer (120) measures battery information from each, for example, of the battery cells in the battery pack (110)).
Accordingly, it would be prima facie obvious to one of ordinary skill in the art, at the time the invention was effectively filed, to have modified Liu with Lee’s measuring a voltage for at least the reasons that a voltage measurement of a particular battery or battery packs may be performed, as taught by Liu; e.g., see para. [0052].
Regarding claim 12, Claim 12 recites The battery capacity estimation method of claim 11, further comprising calculating differential data of the voltage with respect to a time for the battery cell., and is rejected under 35 U.S.C. 103 as being unpatentable by Liu in view of Lee, in further view of Shan for reasons analogous to those set forth in connection with claim 3.
Regarding claim 13, Claim 13 recites The battery capacity estimation method of claim 12, further comprising extracting a plurality of differential data as one principal component data by performing principal component analysis (PCA) on the plurality of differential data. , and is rejected under 35 U.S.C. 103 as being unpatentable by Liu in view of Lee, in further view of Shan for reasons analogous to those set forth in connection with claim 4.
Regarding claim 14, Liu in view of Lee, in further view of Shan discloses The battery capacity estimation method of claim 13, wherein the determining of the second data comprises calculating a plurality of clusters through k-means clustering on the principal component data and determining the differential data as the second data when the differential data is included in a preset cluster. The broadest reasonable interpretation of a method (or process) claim having contingent limitations requires only those steps that must be performed and does not include steps that are not required to be performed because the condition(s) precedent are not met; see, e.g., MPEP 2111.04(III); because the step of calculating a plurality of clusters through k-means clustering on the principal component data and determining the differential data as the second data is only performed if a condition precedent is met; e.g., when the differential data is included in a preset cluster, the broadest reasonable interpretation of this claim does not require this step; e.g., calculating a plurality of clusters through k-means clustering on the principal component data and determining the differential data as the second data to be performed; accordingly, this step does not carry patentable weight.
Regarding claim 15, Liu in view of Lee, in further view of Shan discloses The battery capacity estimation method of claim 14, further comprising determining differential data not included in the cluster as the first data. The broadest reasonable interpretation of a method (or process) claim having contingent limitations requires only those steps that must be performed and does not include steps that are not required to be performed because the condition(s) precedent are not met; see, e.g., MPEP 2111.04(III); because the step of determining of the second data comprises calculating a plurality of clusters through k-means clustering on the principal component data and determining the differential data as the second data is only performed if a condition precedent is met; e.g., when the differential data is included in a preset cluster, the broadest reasonable interpretation of this claim does not require this step; e.g., determining differential data not included in the cluster as the first data to be performed; accordingly, this step does not carry patentable weight.
Claims 3-4 and 12-15 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Lee, in further view of Shan et al. (CN 111178383 A), hereinafter Shan.
Regarding claim 3, Liu in view of Lee discloses The battery capacity estimation apparatus of claim 1, wherein the statistical analyzing unit calculates a time-series voltage curve of the voltage of the battery cell with respect to a time. see rejection as applied to claim 1, specifically fig. 4 illustrating the time-series voltage curves of the majority class, which is cited above as the data determined by the statistical analyzing unit; see also Liu, e.g., see pg. 59006, col. 1 disclosing an example of the process of dividing the discharge voltage time-series dataset into the majority class with the label “1” and the minority class with label “0” is shown in fig. 4. The dividing line is used to redivide the dataset, where the samples below the dividing line are classified as the minority class).
Liu is not relied upon as explicitly disclosing: calculates differential data.
However, Shan further discloses: calculates differential data of the voltage. (Shan, e.g., see pg. 2, lines 40-52 disclosing the charge and discharge when the voltage change rate calculating formula is as follows: monomer wherein vi (k+1) respectively represents the i number of k+1, single voltage of k time, t(k) represents the k-th sampling time; see also pg. 5, lines 10-14 disclosing index data in the step S(100) affects the quality of the battery monomer, comprising: a plurality of voltage change index for extracting sample battery monomer, comprising a charging and discharging when the voltage change rate (every 30 seconds); construed as differential data of the voltage, the monomer voltage and the total average voltage variance, voltage distribution mean, variance of the voltage change, voltage change of the skewness, skewness of the voltage variation distribution two-dimensional vector after dimension reduction; see also pg. 6, lines 34-40 disclosing firstly selecting external quality index influencing battery monomer quality and internal quality index wherein the external quality index for extracting sample battery monomer, comprising charging and discharging when the voltage change rate (every 30 seconds), the variance of the monomer voltage and total average voltage, voltage distribution mean, voltage change, voltage change of the skewness, voltage change distribution of skewness step two from the existing in the data extraction comprises extra-fine, excellent, good, in the difference sample data five-grade quality tag training set; see table of paras. [0080]-[0081]).
Accordingly, it would be prima facie obvious to one of ordinary skill in the art, at the time the invention was effectively filed, to have modified Liu in view of Lee’s apparatus with Shan’s differential data of the voltage for at least the reasons that it is known that voltage differential data identifies peaks in the trace, which indicate battery ageing.
Regarding claim 4, Liu in view of Lee, in further view of Shan is not relied upon as explicitly disclosing: The battery capacity estimation apparatus of claim 3, wherein the statistical analyzing unit extracts a plurality of differential data as one principal component data by performing principal component analysis (PCA) on the plurality of differential data.
However, Shan further discloses: wherein the statistical analyzing unit extracts a plurality of differential data as one principal component data by performing principal component analysis (PCA) on the plurality of differential data. (see rejection as applied to claim 1 disclosing a statistical analyzing unit; see also Shan, e.g., see rejection as applied to claim 3 with regard to pg. 2, lines 40-52, pg. 5, lines 10-14, pg. 6, lines 34-40, and the Table of paras. [0080]-[0081] disclosing differential data to tag a training set, wherein the Table of paras. [0080]-[0081] discloses differential data; see also pg. 6, line 41 – pg. 7, line 7 disclosing the following table contains extra-fine, excellent, good, in the different sample data training set of each grade quality tag after the index to be calculated for each index is transformed into a normalized, then using principal component analysis to internal quality index for the dimension reduction. The internal quality index after standardized processing form a standardized matrix, using principal component analysis to perform dimensionality reduction process, extraction for two of the most important principal component to prevent the different internal index and index number on the support vector machine prediction. The method for calculating the principal component according to claim book in calculating to obtain principal component vector, the vector list display two-dimensional principal component; see also Table of para. [0084] illustrating a first and second principal components; examiner notes principal data 1 or principal data 2 are construed as one principal component data).
Accordingly, it would be prima facie obvious to one of ordinary skill in the art, at the time the invention was effectively filed, to have modified Liu in view of Lee, in further view of Shan’s apparatus with Shan’s statistical analyzing unit extracts a plurality of differential data as one principal component data by performing principal component analysis (PCA) on the plurality of differential data for at least the reasons that principal component analysis utilized in correlation with voltage differentials is known in the art to determine capacity differences in lithium ion batteries due to physics and mathematical properties wherein when a battery loses capacity, the entire differential voltage curve stretches/shrinks in charge direction.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Lee, in further view of Y. Wu, Q. Xue, J. Shen, Z. Lei, Z. Chen and Y. Liu, "State of Health Estimation for Lithium-Ion Batteries Based on Healthy Features and Long Short-Term Memory," in IEEE Access, vol. 8, pp. 28533-28547, 2020, doi: 10.1109/ACCESS.2020.2972344, hereinafter Wu.
Regarding claim 8, Liu in view of Lee is not relied upon as explicitly disclosing: The battery capacity estimation apparatus of claim 1, wherein the voltage measuring unit measures a voltage for a rest period after charging or discharging of the battery cell.
However, Wu further discloses: wherein the voltage measuring unit measures a voltage for a rest period after charging or discharging of the battery cell. (Wu, e.g., see fig. 7 illustrating the evolution trend of HFs [Healthy Features] with cycle numbers; see also pg. 28538, col. 2 – pg. 28539, col. 1 disclosing it can be seen from fig. 7 that the obvious bulges appeared at the 86th,l 101th, 136th, and 186th cycle for
F
5
, which is caused by the rest operation for a period of time between two adjacent cycles).
Accordingly, it would be prima facie obvious to one of ordinary skill in the art, at the time the invention was effectively filed, to have modified Liu in view of Lee’s apparatus with Wu’s voltage measuring unit measures a voltage for a rest period after charging or discharging of the battery cell for at least the reasons that when the cycle experiment is temporarily stopped and the battery remains still for relatively long time, tiny capacity regeneration will occur, as taught by Wu; e.g., see pg. 28539, col. 1.
Claims 5-6 do not stand rejected on the ground(s) of prior art.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
US 2020/0081070 A1 to Chemali et al. relates to a neural-network state-of-charge and state of health estimation.
US 2019/0033395 A1 to Karner et al. systems and methods for monitoring and presenting battery information.
US 2017/0115355 A1 to Willard et al. relates to a maximum capacity estimator for battery state of health and state of charge determinations.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ERIC S. VON WALD whose telephone number is (571)272-7116. The examiner can normally be reached Monday - Friday 7:30 - 5:30.
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, Catherine Rastovski can be reached at (571) 270-0349. 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.
/E.S.V./Examiner, Art Unit 2863
/Catherine T. Rastovski/Supervisory Primary Examiner, Art Unit 2863