DETAILED ACTION
Notice of Pre-AIA or AIA Status
Claims 1-12 are pending in this application.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
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 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.
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
35 U.S.C. § 112 Sixth Paragraph - Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitations are: “unit” in claims 1-12.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
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 may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
Claims 1-12 are rejected under 35 U.S.C. 103 as being unpatentable over Chae et al. (US US20210234207A1, first published July 29, 2021), hereby referred to as “Chae”, in view of Shanghai et al. (Gu Yingzhong, CN113466706B assigned to Shanghai Weixiang Zhongyi New Energy Technology Co ltd, first filed July 26, 2021, first published October 1, 2021), hereby referred to as “Shanghai”. A machine translation of Shanghai was obtained by Google Patents, and a copy is provided herein and was cited for purposes of examination.
Considers Claims 1 and 7.
Chae teaches:
1. A method for detecting an abnormality in battery cell type electrodes, the method comprising: / 7. A device for detecting an abnormality in battery cell type electrodes, the device comprising: (Chae: abstract, A method for analyzing a swelling behavior of a lithium secondary battery includes: (S1) installing a detachable pressurizing holder for fixing a battery sample and a charge/discharge cable connected with an external charging/discharging device to an X-ray CT (computed tomography) imaging equipment; (S2) inserting the battery sample into the pressurizing holder, contacting an electrode lead of the battery sample with a terminal of the charge/discharge cable, and then operating the external charging/discharging device to perform charging and discharging; and (S3) while rotating the battery sample during charging and discharging of the battery sample, irradiating and scanning the X-ray on the battery sample at an angle ranging from −10° to 10° to obtain a 3D image, and then measuring a change in the thickness of the electrode inside the battery sample from the 3D image. [0009]-[0017])
1. acquiring a 3-dimensional image by imaging a specimen including one or more battery cell type electrodes; / 7. an image acquisition unit configured to acquire a 3-dimensional image by imaging a specimen including one or more battery cell type electrodes; (Chae: [0009] The present invention provides a method for analyzing a change in the structure inside an electrode during charging and discharging in real time by establishing a state in which a battery can be charged and discharged in a commercially available X-ray computed tomography (CT) equipment for laboratory (lab), and an analysis system therefor. [0010] An aspect of the present invention provides a method for analyzing a swelling behavior of a lithium secondary battery, the method comprising the steps of: [0011] (S1) installing a detachable pressurizing holder for fixing a battery sample and a charge/discharge cable connected with an external charging/discharging device to an X-ray CT imaging equipment; [0012] (S2) inserting the battery sample into the pressurizing holder, contacting an electrode lead of the battery sample with a terminal of the charge/discharge cable, and then operating the external charging/discharging device to perform charging and discharging; and [0013] (S3) while rotating the battery sample during charging and discharging of the battery sample, irradiating and scanning the X-ray on the battery sample at an angle ranging from −10° to 10° to obtain a 3D image, and then measuring a change in the thickness of the electrode inside the battery sample from the 3D image. [0014] Further, the present invention provides an analysis system for performing the above analysis method, the analysis system comprising: [0015]
(i) a X-ray CT imaging equipment; [0016] (ii) a detachable pressurizing holder installed to the X-ray CT imaging equipment to fix a battery sample; and [0017] (iii) a charge/discharge cable for connecting the battery sample to an external charging/discharging device. [0031]-[0033], Figure 3)
1. determining an arrangement state of the electrodes of the specimen by processing the 3-dimensional image as an input to one or more determination models comprising a rule-based determination model; / 7. an image determination unit configured to determine an arrangement state of the electrodes of the specimen by processing the 3-dimensional image as an input to one or more determination models comprising a rule-based determination model; (Chae: [0031] FIG. 3 is a schematic diagram of a pressurizing holder for fixing a battery sample to an X-ray CT imaging equipment which is used in an analysis method according to an embodiment of the present invention. [0032] Referring to FIG. 3, the pressurizing holder used in the present invention is a device for fixing the battery sample, and includes a pair of pressurizing plate-shaped members 1 for accommodating the battery sample 6, a main screw 2 and auxiliary screws 3 for pressurization, and a window 4 for transmitting the X-ray. The pair of pressurizing plate-shaped members 1 are a structure that faces each other and is spaced apart from each other while the distance therebetween can be adjusted, and may be provided with a pressurizing sponge 5 for preventing a damage to the battery sample 6 between the pair of pressurizing plate-shaped members 1.)
1. and detecting an abnormality in the electrodes of the specimen based on determination results for each of one or more determination models. / 7. and an electrode state determination unit configured to detect an abnormality in the electrodes of the specimen based on determination results for each of one or more determination models. (Chae: [0006] In order to observe a swelling behavior caused by the change in the volume inside the battery according to the charging and discharging conditions, conventionally, an ex-situ method of disassembling a battery cell under a certain SOC (state of charge) condition and then observing the electrode thickness with an electron microscope such as a SEM has been used. FIG. 1 shows a SEM photograph of analyzing a cross-sectional structure of an electrode after sampling a specific cell by disassembling a multi-stacked bi-cell in the conventional ex-situ method. Such an ex-situ disassembling analysis makes impossible to analyze the structure of the entire cell inside a pouch because a maximum width of the bi-cell electrode area that can be observed is limited to within 1 mm, and may cause an error that differs from the actual thickness due to the formation of by-products inside the electrode during the sampling process. [0043] In general, the CT using the X-ray scans 360° to synthesize an image, but after the 360° scan is completed, a change between the captured image and the initial image occurs, making it impossible to align the 3D image. For this reason, it is difficult to observe in real time structural changes within the electrodes included in the battery that occur during the time the scan is completed. Therefore, in order to minimize the image change that occurs during the scan time, it is most accurate to check the change that occurs during the charging/discharging process on a 2D image in which the sample is fixed in a specific direction without rotation, but, with the 2D image that can only be observed in a live form, it is difficult to identify minute structural changes inside the electrode. The 3D scan is essential to overcome these shortcomings, that is, to obtain the 3D image with a higher resolution than the 2D image, and the irradiation and scan angle is advantageously set in the range from −10° to 10° so as to secure the shortest time required for shooting the 3D image. [0054]-[0056], Figures 5-7)
Chae does not teach: one or more determination models comprising a deep learning-based determination model
Shanghai teaches:
1. A method for detecting an abnormality in battery cell type electrodes, the method comprising: / 7. A device for detecting an abnormality in battery cell type electrodes, the device comprising: (Shanghai: abstract, The invention provides a lithium battery echelon utilization residual life prediction method based on a convolutional neural network, which comprises the following steps of: determining the number of required samples according to the relevant battery models; obtaining the battery capacity value and the battery internal resistance value of the training sample by using a constant current and voltage testing method; calculating the residual service life of the lithium battery by taking the internal resistance, the capacity and the charge-discharge cycle curve of the training sample battery as input, and generating a sufficient number of lithium battery service life labels; carrying out X-ray scanning on the lithium battery, and matching the generated image with the service life label to form a training data set; and establishing a echelon battery residual service life model based on the convolutional neural network. According to the lithium battery echelon utilization residual life prediction method based on the convolutional neural network, the convolutional neural network model is established by utilizing the nonlinear relation between the images scanned by the scanning module of the echelon battery and the residual service life, and the residual service life of the lithium battery echelon utilization can be rapidly estimated.)
1. acquiring a 3-dimensional image by imaging a specimen including one or more battery cell type electrodes; / 7. an image acquisition unit configured to acquire a 3-dimensional image by imaging a specimen including one or more battery cell type electrodes; (Shanghai: Summary of the Invention, In order to solve the above-mentioned technical problems, the method for predicting the remaining life of a lithium battery based on a convolutional neural network provided by the present invention includes the following steps:
S1: Use the constant current voltage test method to obtain the battery capacity value and battery internal resistance value of the training sample;
S2: Use the internal resistance, capacity and charge-discharge cycle curve of the training sample battery as input, calculate the remaining service life of the lithium battery, and generate a sufficient number of lithium battery service life labels;
S3: Collect X-ray scan images of lithium batteries, and pair the generated images with service life labels to form a training dataset;
S4: Collect cell images and training samples that identify whether they contain electrodes;
S5: Establish a cell electrode localization model based on convolutional neural network
S6: Establish a model of the remaining service life of the echelon battery based on the convolutional neural network.
S7: Use the established convolutional neural network to perform rapid electrode localization and remaining service life prediction for lithium batteries through X-ray scan images.)
determining an arrangement state of the electrodes of the specimen by processing the 3-dimensional image as an input to one or more determination models comprising a deep learning-based determination model or a rule-based determination model; / 7. an image determination unit configured to determine an arrangement state of the electrodes of the specimen by processing the 3-dimensional image as an input to one or more determination models comprising a deep learning-based determination model or a rule-based determination model; (Shanghai: Summary of the Invention
S4: Collect cell images and training samples that identify whether they contain electrodes;
S5: Establish a cell electrode localization model based on convolutional neural network
S6: Establish a model of the remaining service life of the echelon battery based on the convolutional neural network.
S7: Use the established convolutional neural network to perform rapid electrode localization and remaining service life prediction for lithium batteries through X-ray scan images.
Cell electrode positioning: To estimate the remaining service life of a battery cell by means of image recognition, it is first necessary to locate the position of the electrode of the battery cell by means of machine learning, and then transfer the regional image of the electrode of the battery cell to the next-level deep learning network for prediction of the remaining service life.
Please refer to Figure 4. The initial input is a 256×256RGB color X-ray image, the first hidden layer is a 3×3 convolutional layer with stride 2 and contains 32 channels, and the second hidden layer is a 2×2 step The max pooling layer with length 1 contains 32 channels, the third hidden layer is a 3×3 convolutional layer with stride 2 and contains 64 channels, and the fourth hidden layer is 2×2 with stride 1 A max pooling layer with 64 channels, the fifth hidden layer is a 3×3 convolutional layer with stride 2 with 64 channels, and the sixth hidden layer is 2×2 with stride 1 and contains 64 The max pooling layer of the channel, the linear rectification unit (ReLU) as the activation function of the convolutional layer, has (3×3) kernel size and adopts constant padding. The model is finally connected to 3 fully connected layers with 256, 128 and 64 neurons respectively, and the final output activation function is sigmoid. During the training phase, image patches of the same size are extracted from electrode micrographs, annotated with the correct labels and used to train the CNN model. During the testing phase, a sliding window was used with the trained model to scan photomicrographs of the electrodes. We can control the sensitivity of the model by setting a threshold for recording electrode positioning time. Since the same region containing electrodes may be detected multiple times, we utilize non-maximum suppression (NMS) to ignore redundant, overlapping bounding boxes.)
1. and detecting an abnormality in the electrodes of the specimen based on determination results for each of one or more determination models. / 7. and an electrode state determination unit configured to detect an abnormality in the electrodes of the specimen based on determination results for each of one or more determination models. (Shanghai: Cell electrode positioning: To estimate the remaining service life of a battery cell by means of image recognition, it is first necessary to locate the position of the electrode of the battery cell by means of machine learning, and then transfer the regional image of the electrode of the battery cell to the next-level deep learning network for prediction of the remaining service life.
Please refer to Figure 4. The initial input is a 256×256RGB color X-ray image, the first hidden layer is a 3×3 convolutional layer with stride 2 and contains 32 channels, and the second hidden layer is a 2×2 step The max pooling layer with length 1 contains 32 channels, the third hidden layer is a 3×3 convolutional layer with stride 2 and contains 64 channels, and the fourth hidden layer is 2×2 with stride 1 A max pooling layer with 64 channels, the fifth hidden layer is a 3×3 convolutional layer with stride 2 with 64 channels, and the sixth hidden layer is 2×2 with stride 1 and contains 64 The max pooling layer of the channel, the linear rectification unit (ReLU) as the activation function of the convolutional layer, has (3×3) kernel size and adopts constant padding. The model is finally connected to 3 fully connected layers with 256, 128 and 64 neurons respectively, and the final output activation function is sigmoid. During the training phase, image patches of the same size are extracted from electrode micrographs, annotated with the correct labels and used to train the CNN model. During the testing phase, a sliding window was used with the trained model to scan photomicrographs of the electrodes. We can control the sensitivity of the model by setting a threshold for recording electrode positioning time. Since the same region containing electrodes may be detected multiple times, we utilize non-maximum suppression (NMS) to ignore redundant, overlapping bounding boxes.
PNG
media_image1.png
260
618
media_image1.png
Greyscale
)
It would have been obvious before the effective filing date of the claimed invention was made to one of ordinary skill in the art to modify Chae’s method and system for image analysis of batteries with the teachings of Shanghai for a CNN-based machine learning algorithm for batteries. The determination of obviousness is predicated upon the following findings: both are directed towards methods and systems for image analysis of batteries, and one skilled in the art would have been motivated to modify Chae with the teachings of Shanghai in order to improve the overall image analysis algorithm to leverage a more robust and accurate CNN-based architecture. Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and programming techniques, without changing a “fundamental” operating principle of Chae, while the teaching of Shanghai continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result of a rapid estimation of battery states using image analysis (Shanghai: abstract). It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Considers Claims 2 and 8.
The combination of Chae and Shanghai teaches:
2. The method according to claim 1, wherein in the step of acquiring a 3-dimensional image, the 3-dimensional image is generated by integrating a plurality of images, which are obtained by tomography of cross-sections of the electrodes inside the specimen in a specific region of edges of the specimen, where the electrodes are disposed, by means of CT. / 8. The device according to claim 7, wherein the image acquisition unit generates the 3-dimensional image by integrating a plurality of images, which are obtained by tomography of cross-sections of the electrodes inside the specimen in a specific region of edges of the specimen, where the electrodes are disposed, by means of CT. (Chae: [0006] In order to observe a swelling behavior caused by the change in the volume inside the battery according to the charging and discharging conditions, conventionally, an ex-situ method of disassembling a battery cell under a certain SOC (state of charge) condition and then observing the electrode thickness with an electron microscope such as a SEM has been used. FIG. 1 shows a SEM photograph of analyzing a cross-sectional structure of an electrode after sampling a specific cell by disassembling a multi-stacked bi-cell in the conventional ex-situ method. Such an ex-situ disassembling analysis makes impossible to analyze the structure of the entire cell inside a pouch because a maximum width of the bi-cell electrode area that can be observed is limited to within 1 mm, and may cause an error that differs from the actual thickness due to the formation of by-products inside the electrode during the sampling process. [0054] FIG. 5 shows a cross-sectional image of the electrode obtained by the X-ray scan during charging and discharging of a pouch-shaped battery provided with a Si negative electrode. It can be seen that swelling of about 3% occurs in the electrode according to a charging voltage, which means increase in the electrode thickness. Shanghai First Embodiment, Please refer to Figure 3. The system first disassembles the recovered waste lithium batteries, removes the upper cover of the battery pack, the internal copper insulating cover, removes the wiring harness, battery management system (BMS), electronic components, etc., and disassembles it. It is decomposed into battery modules, and then the battery modules are removed from the upper cover and passed through the lithium battery-specific X-ray scanner in batches by category. The X-ray scan images captured by the scanner will be sent to the cell positioning system based on convolutional neural network. The system is responsible for determining the area of the cell on the X-ray scan image. Then, the sub-image of the cell area is passed into the remaining service life prediction model of the lithium battery for echelon utilization. The model is trained by a large number of historical data of the same type of lithium battery, which can efficiently identify which lithium batteries in the current batch are suitable for echelon utilization and which ones are suitable for echelon utilization. recycle and re-use. The above picture is aimed at the production link of cascade utilization of lithium batteries. If it is a rapid screening of the source of waste lithium batteries, the two steps of disassembling the battery pack and disassembling the battery module can be omitted.)
Considers Claims 3 and 9.
The combination of Chae and Shanghai teaches:
3. The method according to claim 2, wherein the specific region includes a region where an extension line of a first edge formed in a first axis direction and an extension line of a second edge formed in a second axis direction of the specimen meet, and the plurality of images include a first tomography image obtained by tomography of a first cross-section of the specimen perpendicular to the first edge in the specific region, and a second tomography image obtained by tomography of a second cross-section of the specimen perpendicular to the second edge in the specific region. / 9. The device according to claim 8, wherein the specific region includes a region where an extension line of a first edge formed in a first axis direction and an extension line of a second edge formed in a second axis direction of the specimen meet, and the image acquisition unit acquires the plurality of images including a first tomography image obtained by tomography of a first cross-section of the specimen perpendicular to the first edge in the specific region, and a second tomography image obtained by tomography of a second cross-section of the specimen perpendicular to the second edge in the specific region. (Chae: [0035] FIG. 4 schematically shows that the pressurizing holder of FIG. 3 is connected with a charging/discharging device which is located outside a X-ray CT imaging equipment through a charge/discharge cable. [0036] As shown in FIG. 4, the charge/discharge cable 30 connects an electrode lead of the battery sample inserted into the pressurizing holder 20 with the external charging/discharging device 40 inside the X-ray CT imaging equipment 10. Further, the charge/discharge cable 30 may be installed along a high-voltage cable copper line of the X-ray CT imaging equipment so as not to affect shield of the X-ray. For example, the charge/discharge cable may be installed in a zigzag form. Shanghai: Evaluation indicators: The prediction performance of the proposed method for true labels is evaluated using two standard metrics: 1) F1-score, 2) Area under the receiver operating characteristic curve (AUC). Precision(2) is the ratio of the number of true positive TPs divided by the sum of true positive and false positive FPs. It basically describes how well the model predicts the positive class. Recall(3), also known as sensitivity, is the ratio of the number of true positives divided by the sum of true positives and false negatives. The F1-score is the harmonic mean of precision and recall (3). Therefore, it is a more useful metric than the accuracy of uneven class distribution because it takes into account both false positives and false negatives. The Receiver Operating Characteristic (ROC) curve is a plot of false positive rates (x-axis) and true positive rates (y-axis) for various thresholds between 0 and 1. AUC is a useful metric because curves from different models can be directly compared for different thresholds, and the area under the curve can also serve as a summary of the model's predictive performance.
PNG
media_image2.png
822
624
media_image2.png
Greyscale
)
Considers Claims 4 and 10.
The combination of Chae and Shanghai teaches:
4. The method according to claim 1, wherein in the step of determining an arrangement state of the electrodes, each of the deep learning-based determination model and the rule-based determination model is configured to determine an abnormality in the electrodes of the specimen based on at least some of whether an electrode alignment is abnormal, whether an electrode is omitted, whether electrodes are duplicated, and whether an electrode is deformed in the acquired 3-dimensional image. / 10. The device according to claim 7, wherein each of the deep learning-based determination model and the rule-based determination model is configured to determine an abnormality in the electrodes of the specimen based on at least some of whether an electrode alignment is abnormal, whether an electrode is omitted, whether electrodes are duplicated, and whether an electrode is deformed in the acquired 3-dimensional image.(Chae: [0054] FIG. 5 shows a cross-sectional image of the electrode obtained by the X-ray scan during charging and discharging of a pouch-shaped battery provided with a Si negative electrode. It can be seen that swelling of about 3% occurs in the electrode according to a charging voltage, which means increase in the electrode thickness. [0055] FIG. 6 shows thickness measurements on a charge/discharge voltage profile of the Si negative electrode, and FIG. 7 shows a swelling behavior as a change in the thickness according to a charge status (SOC 0% to 100%) of the Si negative electrode. [0056] As can be seen from FIGS. 5 to 7, the swelling behavior of the electrode, that is, a change in the thickness inside the electrode, can be analyzed in real time without disassembling the battery by installing the pressurizing holder for fixing the battery sample and the charge/discharge cable connectable to the external charging/discharging device to the laboratory X-ray CT imaging equipment to establish a state that can perform charging and discharging of the battery sample, and then, irradiating and scanning the X-ray on the battery sample in the range of −10° to 10° while charging and discharging. Shanghai: Cell electrode positioning: To estimate the remaining service life of a battery cell by means of image recognition, it is first necessary to locate the position of the electrode of the battery cell by means of machine learning, and then transfer the regional image of the electrode of the battery cell to the next-level deep learning network for prediction of the remaining service life.
Please refer to Figure 4. The initial input is a 256×256RGB color X-ray image, the first hidden layer is a 3×3 convolutional layer with stride 2 and contains 32 channels, and the second hidden layer is a 2×2 step The max pooling layer with length 1 contains 32 channels, the third hidden layer is a 3×3 convolutional layer with stride 2 and contains 64 channels, and the fourth hidden layer is 2×2 with stride 1 A max pooling layer with 64 channels, the fifth hidden layer is a 3×3 convolutional layer with stride 2 with 64 channels, and the sixth hidden layer is 2×2 with stride 1 and contains 64 The max pooling layer of the channel, the linear rectification unit (ReLU) as the activation function of the convolutional layer, has (3×3) kernel size and adopts constant padding. The model is finally connected to 3 fully connected layers with 256, 128 and 64 neurons respectively, and the final output activation function is sigmoid. During the training phase, image patches of the same size are extracted from electrode micrographs, annotated with the correct labels and used to train the CNN model. During the testing phase, a sliding window was used with the trained model to scan photomicrographs of the electrodes. We can control the sensitivity of the model by setting a threshold for recording electrode positioning time. Since the same region containing electrodes may be detected multiple times, we utilize non-maximum suppression (NMS) to ignore redundant, overlapping bounding boxes.)
Considers Claims 5 and 11.
The combination of Chae and Shanghai teaches:
5. The method according to claim 4, wherein the determination of whether an electrode alignment is abnormal is configured to perform according to: in the step of determining an arrangement state of the electrodes, by using the deep learning-based determination model and the rule-based determination model, whether a gap between endpoints of two electrodes disposed in the 3-dimensional image is greater than a preset reference gap; whether a slope formed based on the endpoints of the two electrodes is greater than a preset reference slope; or whether an endpoint of one electrode from the 3-dimensional image is greater than a preset threshold distance from a preset reference point. / 11. The device according to claim 10, wherein the image determination unit determines whether the electrode alignment is abnormal according to: by using the deep learning-based determination model and the rule-based determination model, whether a gap between endpoints of two electrodes disposed in the 3-dimensional image is greater than a preset reference gap; whether a slope formed based on the endpoints of the two electrodes is greater than a preset reference slope; or whether an endpoint of one electrode from the 3-dimensional image is greater than a preset threshold distance from a preset reference point.(Shanghai: Please refer to Figure 4. The initial input is a 256×256RGB color X-ray image, the first hidden layer is a 3×3 convolutional layer with stride 2 and contains 32 channels, and the second hidden layer is a 2×2 step The max pooling layer with length 1 contains 32 channels, the third hidden layer is a 3×3 convolutional layer with stride 2 and contains 64 channels, and the fourth hidden layer is 2×2 with stride 1 A max pooling layer with 64 channels, the fifth hidden layer is a 3×3 convolutional layer with stride 2 with 64 channels, and the sixth hidden layer is 2×2 with stride 1 and contains 64 The max pooling layer of the channel, the linear rectification unit (ReLU) as the activation function of the convolutional layer, has (3×3) kernel size and adopts constant padding. The model is finally connected to 3 fully connected layers with 256, 128 and 64 neurons respectively, and the final output activation function is sigmoid. During the training phase, image patches of the same size are extracted from electrode micrographs, annotated with the correct labels and used to train the CNN model. During the testing phase, a sliding window was used with the trained model to scan photomicrographs of the electrodes. We can control the sensitivity of the model by setting a threshold for recording electrode positioning time. Since the same region containing electrodes may be detected multiple times, we utilize non-maximum suppression (NMS) to ignore redundant, overlapping bounding boxes.
The training of the model is performed using mini-batch gradient descent, using the Adam optimizer to optimize binary cross-entropy as the loss function (1):
PNG
media_image3.png
183
614
media_image3.png
Greyscale
Shanghai: Evaluation indicators: The prediction performance of the proposed method for true labels is evaluated using two standard metrics: 1) F1-score, 2) Area under the receiver operating characteristic curve (AUC). Precision(2) is the ratio of the number of true positive TPs divided by the sum of true positive and false positive FPs. It basically describes how well the model predicts the positive class. Recall(3), also known as sensitivity, is the ratio of the number of true positives divided by the sum of true positives and false negatives. The F1-score is the harmonic mean of precision and recall (3). Therefore, it is a more useful metric than the accuracy of uneven class distribution because it takes into account both false positives and false negatives. The Receiver Operating Characteristic (ROC) curve is a plot of false positive rates (x-axis) and true positive rates (y-axis) for various thresholds between 0 and 1. AUC is a useful metric because curves from different models can be directly compared for different thresholds, and the area under the curve can also serve as a summary of the model's predictive performance.
PNG
media_image2.png
822
624
media_image2.png
Greyscale
)
Considers Claims 6 and 12.
The combination of Chae and Shanghai teaches:
6. The method according to claim 1, wherein the one or more determination models comprise: i) two or more of the same deep learning-based determination models; ii) two or more different deep learning-based determination models; or iii) at least one of the deep learning-based determination model and at least one rule-based determination model. / 12. The device according to claim 7, wherein the one or more determination models comprises: i) two or more of the same deep learning-based determination models; ii) two or more different deep learning-based determination models; or iii) at least one of the deep learning-based determination model and at least one rule-based determination model. (Chae: [0035] FIG. 4 schematically shows that the pressurizing holder of FIG. 3 is connected with a charging/discharging device which is located outside a X-ray CT imaging equipment through a charge/discharge cable. [0036] As shown in FIG. 4, the charge/discharge cable 30 connects an electrode lead of the battery sample inserted into the pressurizing holder 20 with the external charging/discharging device 40 inside the X-ray CT imaging equipment 10. Further, the charge/discharge cable 30 may be installed along a high-voltage cable copper line of the X-ray CT imaging equipment so as not to affect shield of the X-ray. For example, the charge/discharge cable may be installed in a zigzag form. Shanghai: Please refer to Figure 4. The initial input is a 256×256RGB color X-ray image, the first hidden layer is a 3×3 convolutional layer with stride 2 and contains 32 channels, and the second hidden layer is a 2×2 step The max pooling layer with length 1 contains 32 channels, the third hidden layer is a 3×3 convolutional layer with stride 2 and contains 64 channels, and the fourth hidden layer is 2×2 with stride 1 A max pooling layer with 64 channels, the fifth hidden layer is a 3×3 convolutional layer with stride 2 with 64 channels, and the sixth hidden layer is 2×2 with stride 1 and contains 64 The max pooling layer of the channel, the linear rectification unit (ReLU) as the activation function of the convolutional layer, has (3×3) kernel size and adopts constant padding. The model is finally connected to 3 fully connected layers with 256, 128 and 64 neurons respectively, and the final output activation function is sigmoid. During the training phase, image patches of the same size are extracted from electrode micrographs, annotated with the correct labels and used to train the CNN model. During the testing phase, a sliding window was used with the trained model to scan photomicrographs of the electrodes. We can control the sensitivity of the model by setting a threshold for recording electrode positioning time. Since the same region containing electrodes may be detected multiple times, we utilize non-maximum suppression (NMS) to ignore redundant, overlapping bounding boxes.
The training of the model is performed using mini-batch gradient descent, using the Adam optimizer to optimize binary cross-entropy as the loss function (1):
PNG
media_image3.png
183
614
media_image3.png
Greyscale
Shanghai: Evaluation indicators: The prediction performance of the proposed method for true labels is evaluated using two standard metrics: 1) F1-score, 2) Area under the receiver operating characteristic curve (AUC). Precision(2) is the ratio of the number of true positive TPs divided by the sum of true positive and false positive FPs. It basically describes how well the model predicts the positive class. Recall(3), also known as sensitivity, is the ratio of the number of true positives divided by the sum of true positives and false negatives. The F1-score is the harmonic mean of precision and recall (3). Therefore, it is a more useful metric than the accuracy of uneven class distribution because it takes into account both false positives and false negatives. The Receiver Operating Characteristic (ROC) curve is a plot of false positive rates (x-axis) and true positive rates (y-axis) for various thresholds between 0 and 1. AUC is a useful metric because curves from different models can be directly compared for different thresholds, and the area under the curve can also serve as a summary of the model's predictive performance.
PNG
media_image2.png
822
624
media_image2.png
Greyscale
)
Conclusion
The prior art made of record in form PTO-892 and not relied upon is considered pertinent to applicant's disclosure.
PNG
media_image4.png
186
916
media_image4.png
Greyscale
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TAHMINA ANSARI whose telephone number is 571-270-3379. The examiner can normally be reached on IFP Flex - Monday through Friday 9 to 5.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, O’NEAL MISTRY can be reached on 313-446-4912. The fax phone numbers for the organization where this application or proceeding is assigned are 571-273-8300 for regular communications and 571-273-8300 for After Final communications. TC 2600’s customer service number is 571-272-2600.
Any inquiry of a general nature or relating to the status of this application or proceeding should be directed to the receptionist whose telephone number is 571-272-2600.
2674
/Tahmina Ansari/
January 10, 2026
/TAHMINA N ANSARI/Primary Examiner, Art Unit 2674