Battery cell anomaly detection method, device and equipment and storage medium
By combining principal component analysis and single-class classification techniques with local outlier factors, non-destructive and rapid anomaly detection of lithium battery cells has been achieved, solving the problems of high detection cost and lag in existing technologies, and improving the accuracy and timeliness of detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-09
- Publication Date
- 2026-04-03
Smart Images
Figure CN121784543A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of battery testing technology, and in particular to a method, apparatus, equipment and storage medium for detecting abnormalities in battery cells. Background Technology
[0002] Most methods for detecting anomalies in lithium batteries involve disassembling and analyzing the battery using physical or chemical means to assess its internal state, thereby conducting in-depth research on battery failure mechanisms or verifying non-destructive testing results. However, these methods suffer from high testing costs and testing delays. Summary of the Invention
[0003] In view of this, the purpose of the present invention is to overcome the shortcomings of the prior art and provide a method, device, equipment and storage medium for detecting abnormalities in battery cells, which can achieve non-invasive, efficient and accurate non-destructive identification of potential abnormal states in the operation of battery cells by integrating principal component analysis and single-class classification technology.
[0004] This invention provides the following technical solution: In a first aspect, the present invention proposes a method for detecting abnormalities in battery cells, comprising: Obtain normal electrochemical parameter data and electrochemical parameter data to be analyzed from the battery cell; Principal component analysis was performed based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data; The reconstructed data is classified based on a single-class classification method to obtain abnormal data; The abnormal cell result is determined based on the abnormal data.
[0005] In one embodiment, the step of performing principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data includes: The normal electrochemical parameter data and the electrochemical parameter data to be analyzed are standardized respectively to obtain the first standardized data and the second standardized data. The covariance is calculated on the first standardized data to obtain the covariance matrix; The covariance matrix is decomposed to obtain the principal components and their corresponding eigenvalues; Based on the magnitude of the eigenvalues, the first k principal components are selected to form a dimensionality reduction space; The second standardized data is projected onto the dimensionality reduction space to reconstruct the data, thereby obtaining the reconstructed data.
[0006] In one embodiment, classifying the reconstructed data based on a single-class classification method to obtain abnormal data includes: Calculate the k-nearest neighbor distances corresponding to each reconstructed value in the reconstructed data based on the reconstructed data; Calculate the k-distance neighborhood corresponding to each of the reconstructed values based on the k-nearest neighbor distances corresponding to each of the reconstructed values; Calculate the reachable distance corresponding to each of the reconstructed values based on the k-distance neighborhood corresponding to each of the reconstructed values; Calculate the local reachability density corresponding to each of the reconstructed values based on the reachability distances corresponding to each of the reconstructed values; Calculate the local outlier factor corresponding to each of the reconstructed values based on the local reachability density corresponding to each of the reconstructed values; The anomalous data is obtained based on the local outlier factors corresponding to each of the reconstructed values.
[0007] In one embodiment, obtaining the outlier data based on the local outlier factor includes: For each of the reconstructed values, determine whether the local outlier factor corresponding to the reconstructed value is greater than a preset outlier threshold; If so, then the data corresponding to the reconstructed value is the abnormal data.
[0008] In one embodiment, the standardization of the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain first standardized data and second standardized data includes: The normal electrochemical parameter data and the electrochemical parameter data to be analyzed are respectively transformed into a standard normal distribution to obtain the first standardized data and the second standardized data.
[0009] In one embodiment, determining the cell anomaly result based on the anomaly data includes: Determine the abnormal indicator parameters based on the abnormal data; The abnormal result of the battery cell is determined based on the abnormal indicator parameters.
[0010] In one embodiment, the anomaly index parameters include the number of anomalies, the proportion of anomalies, the anomaly density, and the anomaly clustering situation.
[0011] Secondly, the present invention provides a battery cell anomaly detection device, comprising: The acquisition module is used to acquire normal electrochemical parameter data and electrochemical parameter data to be analyzed from the battery cell. The analysis module is used to perform principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed, so as to obtain reconstructed data; The classification module is used to classify the reconstructed data based on a single-class classification method to obtain abnormal data; The determination module is used to determine the abnormal result of the battery cell based on the abnormal data.
[0012] Thirdly, the present invention proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the cell anomaly detection method as described in the first aspect.
[0013] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the cell anomaly detection method as described in the first aspect.
[0014] This invention discloses a method, apparatus, device, and storage medium for detecting abnormal battery cells. The method acquires normal electrochemical parameter data and electrochemical parameter data to be analyzed from the battery cell. Principal component analysis is performed on the normal and analyzed electrochemical parameter data to obtain reconstructed data. The reconstructed data is then classified using a single-class classification method to obtain abnormal data. Finally, the abnormality of the battery cell is determined based on the abnormal data. In this way, principal component analysis effectively reduces the dimensionality of high-dimensional electrochemical data and extracts features. Combined with single-class classification algorithms such as local outlier factors, unsupervised anomaly identification is achieved. This enables early, rapid, and non-destructive detection of abnormal battery cell states. The entire process does not require disassembling the battery, avoiding the high cost and poor timeliness of traditional physical / chemical detection methods. Attached Figure Description
[0015] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope of protection of the present invention. In the various drawings, similar components are numbered similarly.
[0016] Figure 1 A flowchart of the cell anomaly detection method proposed in this embodiment is shown; Figure 2 A schematic diagram of the battery cell anomaly detection device proposed in this embodiment is shown.
[0017] Explanation of reference numerals in the attached diagram: 200-Cell abnormality detection device; 201-Acquisition module; 202-Analysis module; 203-Classification module; 204-Determination module. Detailed Implementation
[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0019] The components of the embodiments of the invention described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0020] In the following, the terms “comprising,” “having,” and their cognates, which may be used in various embodiments of the invention, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as excluding, firstly, the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations thereof.
[0021] Furthermore, the terms "first," "second," and "third" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.
[0022] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of the invention pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be interpreted as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of the invention.
[0023] Example 1 This disclosure provides a cell anomaly detection method, which integrates principal component analysis and single-class classification technology to achieve non-invasive, efficient, accurate and non-destructive identification of potential abnormal states during cell operation. It is applicable to scenarios such as lithium battery production quality control, battery management system (BMS), and energy storage system health monitoring.
[0024] Please see Figure 1 The cell anomaly detection method includes steps S101 to S104, and each step is described in detail below.
[0025] Step S101: Obtain the normal electrochemical parameter data and the electrochemical parameter data to be analyzed of the battery cell.
[0026] In this embodiment, historical operating data and current data of the battery cell are collected by battery testing equipment or battery management system, thereby obtaining normal electrochemical parameter data and electrochemical parameter data to be analyzed of the battery cell, which serve as the basic input for principal component analysis and anomaly identification.
[0027] Normal electrochemical parameter data refers to a set of multidimensional electrochemical characteristics of a known healthy battery cell, such as quantifiable parameters like voltage, current, temperature, internal resistance, capacity decay rate, charging curve slope, and dQ / dV curve peak position collected under standard charge-discharge cycle conditions. Electrochemical parameter data to be analyzed refers to new rounds of operational data requiring anomaly detection. This data may originate from real-time sampling of the same battery cell during subsequent use, or from test records of other battery cells in the same batch awaiting testing.
[0028] It should be noted that normal electrochemical parameter data should cover typical operating conditions, such as different charge / discharge rates and different temperature environments, to enhance the model's ability to learn normal behavior patterns. The data acquisition frequency can be set according to actual needs, such as once per second or extracting key feature points per charge / discharge cycle.
[0029] Step S102: Perform principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data.
[0030] In this embodiment, principal component analysis is performed using normal electrochemical parameter data to construct a normal coordinate space. This normal coordinate space is then described using the electrochemical parameter data to be analyzed to obtain reconstructed data. At this point, the normal data in the electrochemical parameter data to be analyzed has a smaller reconstruction error based on the principal component direction dominated by the normal electrochemical parameter data, while the abnormal data has a larger reconstruction error, thus providing a basis for subsequent anomaly detection.
[0031] Step S103: Classify the reconstructed data based on a single-class classification method to obtain abnormal data.
[0032] In this embodiment, the reconstructed data is classified based on a single-class classification method, and abnormal data is obtained by filtering. This achieves the goal of not relying on a large number of abnormal samples for training, and is suitable for anomaly detection scenarios where abnormal data is scarce.
[0033] Step S104: Determine the cell abnormality result based on the abnormal data.
[0034] In this embodiment, the abnormal cell result is determined based on abnormal data and in combination with preset rules or a lightweight decision model. The preset rules can be abnormal indicators or weighted scores, thereby improving the accuracy of cell anomaly detection. At the same time, principal component analysis and single-class classification can support online anomaly detection, improve detection timeliness, and avoid the lag in anomaly detection. In addition, principal component analysis and single-class classification avoid damaging the cell core, achieving non-destructive anomaly detection.
[0035] In one specific embodiment, step S102 includes: standardizing the normal electrochemical parameter data and the electrochemical parameter data to be analyzed, respectively, to obtain first standardized data and second standardized data; calculating the covariance of the first standardized data to obtain a covariance matrix; decomposing the covariance matrix to obtain principal components and their corresponding eigenvalues; sorting the eigenvalues according to their magnitude and selecting the first k principal components to form a dimensionality reduction space; and projecting the second standardized data onto the dimensionality reduction space to reconstruct the data and obtain the reconstructed data.
[0036] In this embodiment, normal electrochemical parameter data are standardized to obtain first standardized data; simultaneously, the electrochemical parameter data to be analyzed is standardized to obtain second standardized data, in order to eliminate dimensional errors between parameters. The standardization method can be a standard normal distribution transformation.
[0037] Furthermore, the covariance of the first standardized data is calculated to obtain the covariance matrix, which reflects the linear correlation between the normal electrochemical parameters.
[0038] Furthermore, the covariance matrix is decomposed to obtain the principal components and their corresponding eigenvalues; the first k principal components are selected according to the size of the eigenvalues to form the dimensionality reduction space.
[0039] Furthermore, the second standardized data is projected into the reduced-dimensional space to obtain its score vector along the principal component direction, and then projected back into the original dimensional space to obtain the reconstructed data. This reconstruction process can amplify the reconstruction error of data points that deviate from the normal pattern, thereby providing a sensitive criterion for subsequent anomaly detection.
[0040] In one specific embodiment, step S103 includes: calculating the k-nearest neighbor distance corresponding to each reconstructed value in the reconstructed data based on the reconstructed data; calculating the k-distance neighborhood corresponding to each reconstructed value based on the k-nearest neighbor distance corresponding to each reconstructed value; calculating the reachability distance corresponding to each reconstructed value based on the k-distance neighborhood corresponding to each reconstructed value; calculating the local reachability density corresponding to each reconstructed value based on the local reachability density corresponding to each reconstructed value; and obtaining the anomalous data based on the local outlier factor corresponding to each reconstructed value.
[0041] In this embodiment, the k-nearest neighbor distance corresponding to each reconstructed value in the reconstructed data is calculated based on the reconstructed data, that is, the Euclidean distance between the point and its k-th nearest neighbor.
[0042] Furthermore, the k-distance neighborhood corresponding to each reconstruction value is calculated based on the k-nearest neighbor distance corresponding to each reconstruction value, that is, all points in the neighborhood with the k-nearest neighbor distance as the radius.
[0043] Furthermore, the reachability distance of each reconstructed value relative to its neighbors is calculated based on the k-distance neighborhood corresponding to each reconstructed value.
[0044] Furthermore, the local reachability density corresponding to each reconstructed value is calculated based on the reachability distance corresponding to each reconstructed value, which is the reciprocal of the average reachability distance in its k-neighborhood.
[0045] Furthermore, the local outlier factor corresponding to each reconstructed value is calculated based on the local reachability density corresponding to each reconstructed value. It is defined as the ratio of the local reachability density of the point to the average local reachability density of its k neighboring points.
[0046] Furthermore, outlier data are obtained by filtering out the local outlier factors corresponding to each reconstructed value.
[0047] In one specific embodiment, obtaining the abnormal data based on the local outlier factor includes: for each reconstructed value, determining whether the local outlier factor corresponding to the reconstructed value is greater than a preset outlier threshold; if so, the data corresponding to the reconstructed value is the abnormal data.
[0048] In this embodiment, for each reconstructed value, it is determined whether the local outlier factor corresponding to the reconstructed value is greater than a preset outlier threshold; if so, the data corresponding to the reconstructed value is abnormal data. The preset outlier threshold can be set through historical data verification, cross-validation, or based on experience, and can also be dynamically adjusted according to business needs.
[0049] In one specific embodiment, step S104 includes: determining abnormal indicator parameters based on the abnormal data; and determining the abnormal result of the battery cell based on the abnormal indicator parameters.
[0050] In this embodiment, abnormal indicator parameters are determined based on abnormal data. These parameters may include the number of abnormal points, the proportion of abnormal events, the density of abnormality, and the clustering of abnormality.
[0051] Furthermore, the abnormal results of the battery cell are determined based on the abnormal indicator parameters. For example, if the number of abnormal points exceeds a certain threshold and the proportion of abnormal events is higher than another threshold, it is determined that the battery cell has a risk of performance degradation; if the abnormal points are found to show obvious clustering characteristics, it further indicates that there may be serious defects (such as diaphragm damage or electrode detachment).
[0052] In addition, a graded early warning mechanism can be further combined with detailed analysis of abnormal indicator parameters to conduct graded detection of cell anomalies.
[0053] The proposed cell anomaly detection method in this embodiment acquires normal electrochemical parameter data and electrochemical parameter data to be analyzed from the cell; performs principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data; classifies the reconstructed data based on a single-class classification method to obtain anomaly data; and determines the cell anomaly result based on the anomaly data. In this way, principal component analysis achieves effective dimensionality reduction and feature extraction of high-dimensional electrochemical data, and combined with single-class classification algorithms such as local outlier factors, it completes unsupervised anomaly identification, realizing early, rapid, and non-destructive detection of cell anomalies. The entire process does not require disassembling the battery, avoiding the problems of high cost and poor timeliness associated with traditional physical / chemical detection methods.
[0054] Example 2 Furthermore, this disclosure provides a cell anomaly detection device 200, please refer to [link to relevant documentation]. Figure 2 ,include: The acquisition module 201 is used to acquire normal electrochemical parameter data and electrochemical parameter data to be analyzed of the battery cell; Analysis module 202 is used to perform principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data; Classification module 203 is used to classify the reconstructed data based on a single-class classification method to obtain abnormal data; The determination module 204 is used to determine the abnormal result of the battery cell based on the abnormal data.
[0055] Optionally, the analysis module 202 is further configured to standardize the normal electrochemical parameter data and the electrochemical parameter data to be analyzed, respectively, to obtain first standardized data and second standardized data; calculate the covariance of the first standardized data to obtain a covariance matrix; decompose the covariance matrix to obtain principal components and their corresponding eigenvalues; sort the eigenvalues according to their magnitude and select the first k principal components to form a dimensionality reduction space; project the second standardized data onto the dimensionality reduction space to reconstruct the data and obtain the reconstructed data.
[0056] Optionally, the classification module 203 is further configured to: calculate the k-nearest neighbor distance corresponding to each reconstructed value in the reconstructed data based on the reconstructed data; calculate the k-distance neighborhood corresponding to each reconstructed value based on the k-nearest neighbor distance corresponding to each reconstructed value; calculate the reachability distance corresponding to each reconstructed value based on the k-distance neighborhood corresponding to each reconstructed value; calculate the local reachability density corresponding to each reconstructed value based on the local reachability density corresponding to each reconstructed value; and obtain the abnormal data based on the local outlier factor corresponding to each reconstructed value.
[0057] Optionally, the classification module 203 is further configured to determine, for each of the reconstructed values, whether the local outlier factor corresponding to the reconstructed value is greater than a preset outlier threshold; if so, the data corresponding to the reconstructed value is the abnormal data.
[0058] Optionally, the analysis module 202 is further configured to perform standard normal distribution transformation on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed, respectively, to obtain the first standardized data and the second standardized data.
[0059] Optionally, the determining module 204 is further configured to determine abnormal indicator parameters based on the abnormal data; and determine the abnormal result of the battery cell based on the abnormal indicator parameters.
[0060] Optionally, the anomaly index parameters include the number of anomalies, the proportion of anomalies, the anomaly density, and the anomaly clustering situation.
[0061] The apparatus provided in this embodiment can perform the steps of the cell anomaly detection method provided in Embodiment 1. To avoid repetition, the steps will not be repeated.
[0062] The battery cell anomaly detection device proposed in this embodiment acquires normal electrochemical parameter data and electrochemical parameter data to be analyzed from the battery cell; performs principal component analysis on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data; classifies the reconstructed data based on a single-class classification method to obtain anomaly data; and determines the battery cell anomaly result based on the anomaly data. In this way, principal component analysis achieves effective dimensionality reduction and feature extraction of high-dimensional electrochemical data, and combined with single-class classification algorithms such as local outlier factors, it completes unsupervised anomaly identification, realizing early, rapid, and non-destructive detection of battery cell anomalies. The entire process does not require disassembling the battery, avoiding the problems of high cost and poor timeliness associated with traditional physical / chemical detection methods.
[0063] Example 3 Furthermore, this disclosure provides an electronic device including a memory and a processor. The memory stores a computer program, which, when executed by the processor, implements the cell anomaly detection method described in Embodiment 1.
[0064] The device provided in this embodiment can perform the steps of the cell anomaly detection method provided in Embodiment 1. To avoid repetition, the steps will not be repeated.
[0065] Example 4 This disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the cell anomaly detection method described in Embodiment 1.
[0066] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.
[0067] The computer-readable storage medium provided in this embodiment can implement the cell anomaly detection method provided in Embodiment 1. To avoid repetition, it will not be described again here.
[0068] In all examples shown and described herein, any specific values should be interpreted as merely exemplary and not as limitations; therefore, other examples of exemplary embodiments may have different values.
[0069] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.
[0070] The above-described embodiments are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.
Claims
1. A method for detecting abnormalities in battery cells, characterized in that, include: Obtain normal electrochemical parameter data and electrochemical parameter data to be analyzed from the battery cell; Principal component analysis was performed based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data; The reconstructed data is classified based on a single-class classification method to obtain abnormal data; The abnormal cell result is determined based on the abnormal data.
2. The cell anomaly detection method according to claim 1, characterized in that, The step of performing principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed to obtain reconstructed data includes: The normal electrochemical parameter data and the electrochemical parameter data to be analyzed are standardized respectively to obtain the first standardized data and the second standardized data. The covariance is calculated on the first standardized data to obtain the covariance matrix; The covariance matrix is decomposed to obtain the principal components and their corresponding eigenvalues; Based on the magnitude of the eigenvalues, the first k principal components are selected to form a dimensionality reduction space; The second standardized data is projected onto the dimensionality reduction space to reconstruct the data, thereby obtaining the reconstructed data.
3. The cell anomaly detection method according to claim 1, characterized in that, The reconstructed data is classified using a single-class classification method to obtain abnormal data, including: Calculate the k-nearest neighbor distances corresponding to each reconstructed value in the reconstructed data based on the reconstructed data; Calculate the k-distance neighborhood corresponding to each of the reconstructed values based on the k-nearest neighbor distances corresponding to each of the reconstructed values; Calculate the reachable distance corresponding to each of the reconstructed values based on the k-distance neighborhood corresponding to each of the reconstructed values; Calculate the local reachability density corresponding to each of the reconstructed values based on the reachability distances corresponding to each of the reconstructed values; Calculate the local outlier factor corresponding to each of the reconstructed values based on the local reachability density corresponding to each of the reconstructed values; The anomalous data is obtained based on the local outlier factors corresponding to each of the reconstructed values.
4. The cell anomaly detection method according to claim 3, characterized in that, The process of obtaining the outlier data based on the local outlier factor includes: For each of the reconstructed values, determine whether the local outlier factor corresponding to the reconstructed value is greater than a preset outlier threshold; If so, then the data corresponding to the reconstructed value is the abnormal data.
5. The cell anomaly detection method according to claim 2, characterized in that, The standardization of the normal electrochemical parameter data and the electrochemical parameter data to be analyzed, respectively, to obtain first standardized data and second standardized data, includes: The normal electrochemical parameter data and the electrochemical parameter data to be analyzed are respectively transformed into a standard normal distribution to obtain the first standardized data and the second standardized data.
6. The cell anomaly detection method according to claim 1, characterized in that, The step of determining the cell anomaly result based on the abnormal data includes: Determine the abnormal indicator parameters based on the abnormal data; The abnormal result of the battery cell is determined based on the abnormal indicator parameters.
7. The cell anomaly detection method according to claim 6, characterized in that, The abnormality index parameters include the number of abnormal points, the proportion of abnormal events, the abnormal density, and the abnormal clustering situation.
8. A battery cell anomaly detection device, characterized in that, include: The acquisition module is used to acquire normal electrochemical parameter data and electrochemical parameter data to be analyzed from the battery cell. The analysis module is used to perform principal component analysis based on the normal electrochemical parameter data and the electrochemical parameter data to be analyzed, so as to obtain reconstructed data; The classification module is used to classify the reconstructed data based on a single-class classification method to obtain abnormal data; The determination module is used to determine the abnormal result of the battery cell based on the abnormal data.
9. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, implements the cell anomaly detection method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the cell anomaly detection method as described in any one of claims 1 to 7.