Methods, systems, equipment, media, and products for locating abnormal batteries

Abstract

This application relates to a method, system, device, medium, and product for locating abnormal batteries. The method for locating abnormal batteries includes: determining the battery pack's operational data; performing outlier detection and normalization on the operational data to obtain target operational data; performing time series analysis on the target operational data to determine temporal characteristics, and obtaining spatial characteristics based on the location information of each battery and the target operational data; inputting the spatiotemporal characteristics into a thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal characteristics; and determining that the battery corresponding to the target area is abnormal in response to the target feature value in the thermal image exceeding a preset threshold. This method can solve the problems of traditional infrared thermal imaging technology, such as being greatly affected by the environment, limited installation space, high cost, and insufficient spatiotemporal resolution of weak signals. It can accurately reflect the thermal state of the battery pack, achieving early abnormal location and accurate early warning of thermal runaway.

Description

Technical Field

[0001] This application relates to the field of battery safety monitoring technology, and in particular to a method, system, device, medium, and product for locating abnormal batteries. Background Technology

[0002] In battery applications, abnormal conditions such as overheating, overcharging, and short circuits in battery packs can lead to thermal runaway. This process is often accompanied by a rapid rise in temperature, gas release, and even electrolyte combustion, which can easily cause catastrophic fires or explosions. Traditional infrared thermal imaging technology, which relies on surface temperature detection, is easily affected by changes in ambient temperature and humidity, light interference, and physical barriers from the battery casing. This results in an inability to accurately reflect the thermal state of the battery pack. Furthermore, it requires the installation of a thermal imager, which is space-constrained and costly. In addition, in the early stages of thermal runaway, the weak signals of localized minute temperature differences and hidden heat accumulation cannot be monitored and imaged in real time. Summary of the Invention

[0003] Based on this, a method, system, device, medium, and product for locating abnormal batteries are provided, which solves the problem that traditional thermal imaging technology cannot accurately monitor the thermal state of battery packs in the prior art.

[0004] Firstly, a method for locating an abnormal battery is provided, the method comprising:

[0005] The operating data of the battery pack is determined; wherein, the operating data includes surface temperature, internal temperature, gas composition, gas concentration, and casing deformation pressure, and the operating data is obtained based on a sensor array disposed on the battery pack; the sensor array includes a flexible thin-film temperature sensor, a distributed fiber optic temperature sensor, a gas sensor, a voltage sensor, and a micro-differential pressure sensor; the battery pack includes at least two batteries;

[0006] The running data is subjected to outlier detection and normalization to obtain the target running data;

[0007] Time series analysis is performed on the target operation data to determine time features, and spatial features are obtained based on the location information of each battery and the target operation data. Attention scores of the time features and spatial features are calculated through linear transformation. The attention scores are normalized to the weights of the time features and spatial features using a predetermined function. The time features and spatial features are then weighted and summed according to the normalized weights to obtain the spatiotemporal features.

[0008] The spatiotemporal features are input into the thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal features; the spatiotemporal features include the temporal features and the spatial features.

[0009] In response to the target feature value of the target area in the thermal image exceeding a preset threshold, the battery anomaly corresponding to the target area is determined; the target feature value includes temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change amplitude, and pressure change rate.

[0010] Optionally, the temperature sensor includes at least one of the following: a flexible thin-film temperature sensor, a distributed optical fiber temperature sensor, a thermocouple sensor, or a thermistor; the flexible thin-film temperature sensor is embedded in the gaps between the batteries in a matrix form, and the distributed optical fiber temperature sensor is arranged in a serpentine pattern around the battery pack.

[0011] Optionally, the outlier detection method includes at least one of the following: Z-score method, IQR method, and isolated forest algorithm;

[0012] The normalization process includes at least one of the following: min-max normalization and Z-score normalization.

[0013] Optionally, the predetermined function includes the Softmax function.

[0014] Optionally, obtaining the spatial features based on the location information of each battery and the target operating data includes:

[0015] Construct a spatial coordinate matrix based on the physical location of each battery in the battery pack;

[0016] The spatial features are obtained based on the spatial coordinate matrix and the target running data.

[0017] Optionally, the positioning method further includes:

[0018] Calculate the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas in the image region corresponding to each battery in the thermal imaging image;

[0019] Based on the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas, the risk level of the battery corresponding to the image area is determined.

[0020] In response to any image region having a risk level exceeding a preset level, a battery anomaly is determined corresponding to that image region.

[0021] Secondly, a system for locating abnormal batteries is provided, the system comprising:

[0022] The first determining module is used to determine the operating data of the battery pack; wherein, the operating data includes surface temperature, internal temperature, gas composition, gas concentration, and casing deformation pressure, and the operating data is obtained based on a sensor array disposed on the battery pack; the sensor array includes a flexible thin-film temperature sensor, a distributed fiber optic temperature sensor, a gas sensor, a voltage sensor, and a micro-differential pressure sensor; the battery pack includes at least two batteries.

[0023] The processing module is used to perform outlier detection and normalization on the running data to obtain the target running data;

[0024] The second determining module is used to perform time series analysis on the target operation data to determine time characteristics, and obtain spatial characteristics based on the location information of each battery and the target operation data;

[0025] An image generation module is used to input spatiotemporal features into a thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal features; the spatiotemporal features include the temporal features and the spatial features;

[0026] The third determining module is used to determine the battery anomaly corresponding to the target area in response to the target feature value of the target area in the thermal imaging exceeding a preset threshold; the target feature value includes temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change amplitude, and pressure change rate.

[0027] Thirdly, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and for running on the processor, wherein the processor executes the computer program to implement the method of the first aspect.

[0028] Fourthly, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0029] Fifthly, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in the first aspect.

[0030] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this application.

[0031] The aforementioned method, system, equipment, medium, and product for locating abnormal batteries obtain target operational data of the batteries through a sensor array mounted on the battery pack. Based on the target operational data of the battery pack and the position information of each battery within the pack, spatiotemporal characteristics are determined, and a thermal image of the battery pack is generated based on these characteristics. When the target feature value of a target area in the thermal image exceeds a preset threshold, the battery corresponding to that target area is identified as abnormal. This method solves the problems of traditional infrared thermal imaging technology, such as significant environmental influences, limited installation space, high cost, and insufficient spatiotemporal resolution of weak signals. It can accurately reflect the thermal state of the battery pack, enabling early abnormal location and precise early warning of thermal runaway. Attached Figure Description

[0032] Figure 1 This is a flowchart of a method for locating an abnormal battery in one embodiment;

[0033] Figure 2 This is a schematic diagram of sensor connections for a method of locating an abnormal battery in one embodiment;

[0034] Figure 3 This is a schematic diagram of the data preprocessing flow for a method of locating abnormal batteries in one embodiment;

[0035] Figure 4 This is a schematic diagram of the spatiotemporal feature extraction process of an abnormal battery localization method in one embodiment;

[0036] Figure 5 This is a schematic diagram of the thermal imaging spatial output of a method for locating abnormal batteries in one embodiment;

[0037] Figure 6 This is a flowchart illustrating the abnormal battery location method in one embodiment.

[0038] Figure 7 This is a schematic diagram of the structure of a faulty battery location system in one embodiment;

[0039] Figure 8 This is a schematic diagram of the structure of an electronic device in one embodiment. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0041] It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of this application. Therefore, the drawings only show components relevant to this application and are not drawn according to the actual number, shape, and size of components in implementation. In actual implementation, the form, quantity, and proportion of each component can be arbitrarily changed, and the component layout may also be more complex. The structures, proportions, sizes, etc., shown in the accompanying drawings are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the implementation conditions of this application. Therefore, they have no substantial technical significance. Any modification to the structure, change in the proportional relationship, or adjustment of the size, without affecting the effect and purpose that this application can produce, should still fall within the scope of the technical content disclosed in this application. At the same time, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are only for clarity of description and are not intended to limit the scope of implementation of this application. Changes or adjustments in their relative relationships, without substantially changing the technical content, should also be considered within the scope of implementation of this application.

[0042] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the document does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0043] As illustrated herein, unless the context clearly indicates otherwise, the words “a,” “an,” “an,” and / or “the” do not specifically refer to the singular and may also include the plural. Generally speaking, the terms “comprising” and “including” only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0044] The definitions used herein, such as the terms “having,” “may have,” “comprising,” or “may include,” indicate the presence of the corresponding function, operation, element, etc., and do not limit the presence of one or more other functions, operations, elements, etc. Furthermore, it should be understood that the terms “comprising” or “having” as used herein indicate the presence of the features, figures, steps, operations, elements, components, or combinations thereof described in the specification, without excluding the presence or addition of one or more other features, figures, steps, operations, elements, components, or combinations thereof.

[0045] The prefixes such as "first" and "second" used in this application embodiment are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this application embodiment does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not constitute unnecessary restrictions due to the use of such prefixes. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0046] In battery applications, abnormal conditions such as overheating, overcharging, and short circuits can trigger thermal runaway. This process is often accompanied by a rapid rise in temperature, gas release, and even electrolyte combustion, which can easily lead to catastrophic fires or explosions. Traditional infrared thermal imaging technology has three major drawbacks: 1. Physical limitations: It requires direct viewing of the battery surface and cannot penetrate the casing to monitor the internal thermal state; 2. Environmental interference: Changes in ambient temperature and humidity cause surface temperature measurement errors > ±3℃; 3. Insufficient dynamic response: The frame rate is usually ≤10Hz, making it difficult to capture the millisecond-level temperature rise in the early stages of thermal runaway. Early warning systems for battery thermal runaway can accurately identify potential thermal runaway risks by monitoring multimodal characteristics such as battery temperature, voltage, pressure, and gas in the early stages of an incident. Feature imaging, as a visualization technology, is widely used in industrial non-destructive testing, and it is particularly valuable in the perception and identification of complex systems such as lithium-ion battery packs. Therefore, researching thermal imaging methods based on the spatiotemporal characteristics of battery packs is of great significance for the safety monitoring and early warning of thermal runaway in energy storage battery packs. Based on this, this application provides a method, system, device, medium, and product for locating abnormal batteries, which solves the problems of traditional infrared thermal imaging technology being greatly affected by the environment, having limited installation space, high cost, and insufficient spatiotemporal resolution of weak signals. It can accurately reflect the thermal state of the battery pack and realize early abnormal location and accurate early warning of thermal runaway.

[0047] Figure 1 The present application provides a method for locating abnormal batteries, which can be used to locate abnormal batteries in battery packs, such as abnormal batteries in electric vehicle power batteries and energy storage battery packs in energy storage power stations. The location method includes:

[0048] S11. Determine the battery pack's operating data.

[0049] The operating data includes the battery pack's surface temperature, internal temperature, gas composition, gas concentration, and casing deformation pressure.

[0050] Operating data can be obtained through a sensor array mounted on the battery pack. This sensor array includes flexible temperature sensors, gas sensors, voltage sensors, differential pressure sensors, and pressure sensors. Figure 2As shown, the temperature sensor includes at least one of the following: a flexible thin-film temperature sensor, a distributed fiber optic temperature sensor, a thermocouple sensor, or a thermistor. For example, the temperature sensor is installed on the surface or inside the battery pack and physically connected to the battery pack body to accurately obtain the surface and internal temperatures of the battery. A voltage sensor can be located between the positive and negative terminals of the battery pack or directly connected to the positive and negative terminals, and electrically connected to the battery pack, to measure the battery's terminal voltage, reflecting the battery's charging and discharging state. A micro-differential pressure sensor is located inside the battery pack to detect and monitor the composition and concentration of gases released by the battery. A casing deformation pressure indicator indicates the pressure value when the battery casing deforms due to increased internal pressure during operation; a pressure sensor can be located on the surface of the battery pack and physically connected to the battery pack body to measure the deformation pressure of the battery pack casing. This multimodal data can comprehensively reflect the battery's state, providing rich information for subsequent analysis.

[0051] In one embodiment, flexible thin-film temperature sensors are embedded in the gaps between batteries in a matrix form, and distributed fiber optic temperature sensors are arranged in a serpentine pattern around the battery pack.

[0052] Flexible thin-film temperature sensors possess excellent flexibility, capable of deforming and tightly conforming to various curved and irregularly shaped objects. Embedded in a matrix within the gaps between batteries, these sensors rapidly detect temperature changes with a short response time, typically outputting accurate temperature signals within seconds or even less. Sensitive to environmental changes, they accurately reflect ambient temperature variations, making them suitable for temperature monitoring in various complex environments. Distributed fiber optic temperature sensors, on the other hand, are immune to electromagnetic interference, making them suitable for strong electromagnetic fields. They eliminate safety hazards such as leakage and electric shock, and are free from external electromagnetic interference, making them ideal for use in electromagnetically sensitive environments. They can achieve millimeter-level or even higher spatial resolution. The combined layout of flexible thin-film and distributed fiber optic temperature sensors overcomes the dependence of traditional infrared thermal imaging on battery surface temperature.

[0053] S12. Perform outlier detection and normalization on the running data to obtain the target running data.

[0054] The collected operational data undergoes preprocessing, including outlier detection and normalization, to obtain the target operational data. Outlier detection removes noise and erroneous data, ensuring the reliability of the operational data, while normalization converts data with different dimensions into a unified dimension, facilitating subsequent analysis and processing.

[0055] In one embodiment, the outlier detection method includes at least one of the following: Z-score method, IQR method, and Isolation Forest algorithm;

[0056] The normalization process includes at least one of the following: min-max normalization and Z-score normalization.

[0057] Outlier detection can employ statistical or machine learning methods, such as the Z-score method, IQR (Interquartile Range Method), or the Isolation Forest algorithm. Normalization can be achieved using min-max normalization or Z-score normalization to suit different data characteristics.

[0058] For example, the target operating data includes multimodal data such as the battery pack's surface temperature, internal temperature, gas composition, gas concentration, and casing deformation pressure. Figure 3 As shown, outlier detection and normalization were performed on the collected multimodal data. Outlier detection employed the statistically based Z-score method, calculating the deviation of each data point from the mean. Data points exceeding a certain standard deviation were considered outliers and corrected or removed. Normalization used the min-max normalization method to transform the data to the [0,1] interval, as shown in the formula:

[0059] ;

[0060] Among them, X n The data is normalized, and X is the original data. min and X max These are the minimum and maximum values ​​of the original data, respectively.

[0061] S13. Perform time series analysis on the target operation data to determine the time characteristics, and obtain the spatial characteristics based on the location information of each battery and the target operation data.

[0062] Time series analysis can employ various models, such as Autoregressive Model (AR), Moving Average Model (MA), Autoregressive Moving Average Model (ARMA), Autoregressive Integrated Moving Average Model (ARIMA), or Long Short-Term Memory Network (LSTM), selecting the appropriate analysis method based on the temporal characteristics of the data.

[0063] By using time series analysis and spatial feature construction, the spatiotemporal characteristics of thermal runaway of the battery pack can be extracted. Time series analysis can uncover the changing patterns of data in the time dimension, while spatial feature construction can consider the physical positional relationship of each battery in the battery pack, thus comprehensively considering the characteristics in both time and space dimensions.

[0064] In one embodiment, obtaining spatial features based on the location information of each battery and the target operation data includes: constructing a spatial coordinate matrix based on the physical location of each battery in the battery pack; and obtaining spatial features based on the spatial coordinate matrix and the target operation data.

[0065] like Figure 4 As shown, time series analysis can be performed on the preprocessed target operating data using a Long Short-Term Memory (LSTM) network to extract temporal features. LSTM can capture dependencies in long time series and is suitable for processing the temporal variation patterns of battery state data. Simultaneously, a spatial coordinate matrix is ​​constructed based on the physical location of each battery within the battery pack, with each battery's position represented by (x, y, z) coordinates. Combined with the target operating data, a spatial feature vector is generated. For example, temperature, voltage, pressure, and gas data are fused with spatial coordinates to obtain a feature vector containing spatial location information, i.e., spatial features.

[0066] In one embodiment, after performing time series analysis on the target operation data to determine temporal characteristics and obtaining spatial characteristics based on the location information of each battery and the target operation data, the method further includes:

[0067] Based on the attention mechanism, weights are assigned to temporal and spatial features respectively to obtain spatiotemporal features.

[0068] After determining the temporal and spatial features, attention scores for the temporal and spatial features can be calculated using linear transformation. The attention scores are then normalized to the weights of the temporal and spatial features using the Softmax function. Finally, the temporal and spatial features are weighted and summed based on the normalized weights to obtain the spatiotemporal features.

[0069] S14. Input the spatiotemporal features into the thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal features.

[0070] Spatiotemporal characteristics include temporal characteristics and spatial characteristics.

[0071] A pre-trained thermal imaging model can be used, employing machine learning algorithms to construct the model using spatiotemporal features as input, mapping these features to the thermal imaging space. Machine learning algorithms possess powerful nonlinear mapping capabilities, accurately establishing the relationship between spatiotemporal features and thermal imaging. Model training methods include:

[0072] 1) Construction of training dataset

[0073] Data source: Multimodal data of lithium battery packs under normal operation, overcharge, overheating, short circuit and other conditions are collected, including several groups of tagged samples (tag definition: normal state / warning state / thermal runaway state).

[0074] Enhancement strategies: Add Gaussian noise (σ=0.05, where σ indicates standard deviation) to the time series data and randomly flip / rotate the spatial grid data to improve the model's generalization ability.

[0075] 2) Loss Function Design

[0076] Pixel-level regression loss: The pixel difference between the predicted thermal image and the actual thermal image acquired by the infrared thermal imager is calculated using mean squared error (MSE).

[0077] State classification loss: The cross-entropy loss function is combined to optimize the classification results of the warning state, with weight coefficients α=0.7 (α indicates regression loss); β=0.3 (β indicates classification loss).

[0078] 3) Optimization Algorithm

[0079] The AdamW (Adam with Weight Decay) optimizer (learning rate 1e-4, weight decay 0.01) was used, along with a cosine annealing learning rate scheduling strategy, and the training period was set to 50 epochs.

[0080] Machine learning algorithms can include: Support Vector Machine (SVM), Random Forest, Convolutional Neural Network (CNN), or Generative Adversarial Network (GAN), etc.

[0081] After constructing the thermal imaging model, the spatiotemporal features are input into the thermal imaging model to generate a thermal image of the battery pack, including the following steps:

[0082] (1) Spatiotemporal feature coding layer

[0083] Input data format: Convert the spatiotemporal feature vector into a three-dimensional tensor [number of batteries, time series length, feature dimension], where the feature dimension contains temperature / voltage / pressure / gas concentration data and their spatial coordinates (x, y, z).

[0084] Location embedding technology: The physical location (x, y, z) of the battery is encoded by sine and cosine functions to generate a location embedding vector, which is added element by element to the sensor data features to achieve deep fusion of spatial location information and state data.

[0085] (2) Temporal Feature Extraction Network

[0086] Main network structure: A two-layer bidirectional long short-term memory network (Bi-LSTM) is used, with 128 neurons in the hidden layer and a dropout layer (with a retention rate of 0.5) to prevent overfitting.

[0087] Output features: Extract dynamic features of the time series, including the rate of temperature change (dT / dt) and voltage fluctuation coefficient. Derivative features such as pressure gradient (dP / dt) are also included.

[0088] (3) Spatial Feature Convolutional Network

[0089] The 2D convolution module converts the physical layout of the battery pack into an N×M grid matrix (N is the number of rows and M is the number of columns), and each grid node loads the preprocessed data of the corresponding battery.

[0090] Convolution kernel design: A 3×3 local receptive field convolution kernel is used, combined with dilated convolution (dilation rate=2) to expand the spatial perception range and extract the thermal conduction correlation features between adjacent cells.

[0091] Pooling strategy: Alternate between max pooling (preserving spatial anomalies) and average pooling (capturing global heat distribution trends).

[0092] (4) Feature fusion and mapping layer

[0093] Cross-modal fusion: The temporal features (dimension 1×256) output by Bi-LSTM (Bidirectional Long Short-Term Memory) and the spatial features (dimension N×M×64) extracted by CNN are dimensionally aligned to form a spatiotemporal joint feature tensor by tensor concatenation.

[0094] Nonlinear mapping: A multi-layer fully connected network (2 layers, 256 neurons per layer, activation function ReLU (Rectified Linear Unit)) is used to map the joint features to the thermal imaging space, and the output is an N×M×3 RGB pixel matrix (corresponding to the red, green and blue channel values ​​of the thermal image).

[0095] For example, Figure 5As shown, a thermal imaging model can be constructed using a convolutional neural network (CNN) with spatiotemporal features as input. CNNs have powerful image processing and feature extraction capabilities, enabling the mapping of spatiotemporal features to the thermal imaging space. The spatiotemporal feature vectors are converted into a matrix form suitable for CNN input, and through processing by convolutional layers, pooling layers, and fully connected layers, the pixel value matrix of the thermal image is output.

[0096] After generating a thermal image of the battery pack, the thermal distribution of the battery pack can be seen intuitively, allowing for the timely detection of potential thermal runaway risks, such as... Figure 6 As shown:

[0097] (1) Thermal image generation technology

[0098] 1) Pseudo-color mapping algorithm

[0099] Temperature-color mapping is achieved using the HSV color space conversion model.

[0100] Hue (H): 0°-180° corresponds to a gradient from blue (cool) to red (warm);

[0101] Saturation (S): Fixed at 90%, to enhance color differentiation;

[0102] Brightness (V): Dynamically adjusted according to temperature value, V=30% in low temperature zone and V=90% in high temperature zone.

[0103] 2) Rules for marking abnormal areas

[0104] Warning area (temperature rise rate ≥ 2℃ / s): Add a yellow dashed box around the edge;

[0105] Early thermal runaway region (multiple features exceeding limits): Displayed using a red checkerboard texture.

[0106] 3) Multi-dimensional visualization output

[0107] Two-dimensional planar view: The physical layout of the battery pack is used as a grid, and each cell displays the thermal state value (combined temperature / voltage / pressure value) and color mapping result of the corresponding battery;

[0108] 3D stereoscopic view: Construct a 3D coordinate system with xy position axis and z temperature axis, generate heat distribution surface map through surface interpolation algorithm, and support mouse drag and rotation to view any angle.

[0109] S15. In response to the target feature value of the target area in the thermal imaging exceeding a preset threshold, determine the battery anomaly corresponding to the target area.

[0110] Using a pre-constructed thermal imaging model, a thermal image of the battery pack is generated. Each pixel in the thermal image corresponds to a location within the battery pack, and the pixel's color indicates the thermal state at that location. Based on preset early warning thresholds for thermal runaway, such as temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change magnitude, and pressure change rate, abnormal cells in the battery pack are located and their thermal states are assessed. When features in certain areas exceed the thresholds, an alarm is triggered for the abnormal cells, alerting relevant personnel to take appropriate action.

[0111] Once a thermal image is generated, abnormal battery conditions can be identified, for example, by setting a temperature threshold T. th =60℃, temperature change rate threshold ΔT / Δt =5℃ / s, ΔT indicates the temperature change, Δt indicates the time change. When a battery cell in the reconstructed thermal image exceeds the threshold for three consecutive time points, and the gas concentration at the corresponding location is abnormal ( If the battery is identified as abnormal, its coordinates and risk level will be output.

[0112] In one embodiment, the positioning method further includes:

[0113] Calculate the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas in the image region corresponding to each battery in the thermal image.

[0114] The risk level of the battery corresponding to the image region is determined based on the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas.

[0115] If the risk level of any image region exceeds a preset level, the battery anomaly corresponding to that image region is determined.

[0116] Using a pre-constructed thermal imaging model, a thermal image of the battery pack is generated. Each pixel in the thermal image corresponds to a location within the battery pack, and the pixel's color indicates the thermal state at that location. Based on preset early warning thresholds for thermal runaway, such as temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change amplitude, and pressure change rate, the abnormal cells in the battery pack are located and their thermal states are assessed. The temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas corresponding to each cell in the thermal image are calculated. Based on these factors, the risk level of the cell corresponding to the image area is determined. When the risk level of any image area exceeds a preset level, an alarm is triggered for the abnormal cell, alerting relevant personnel to take appropriate measures.

[0117] Among them, the temperature rise rate indicates the speed at which the temperature increases over time, and can be measured in °C / min or °C / s; the temperature mutation index indicates the degree of drastic temperature change in a short period of time; and the gas correlation indicates the ratio or change relationship between the concentrations of different characteristic gases. When an abnormal chemical reaction occurs inside the battery, trace amounts of characteristic gases, such as small molecule gases like H2, CO, and C2H4, will be preferentially generated. The gas correlation can include the H2 / CO ratio, The ratio, the rate of change of total hydrocarbon concentration, etc., therefore, compared with the single gas concentration information, the gas correlation can amplify the small chemical reactions generated inside the battery and provide clearer and earlier battery abnormal signals; the voltage abnormal change amplitude indicates the degree to which the voltage deviates from the normal range in a short period of time, which can be determined by measuring the voltage fluctuation range; the pressure change rate indicates the rate at which the pressure changes over time.

[0118] Temperature gradients indicate the rate of temperature change in space, representing the degree of temperature variation per unit distance. Batteries generate heat during charging and discharging, and temperature gradients help identify uneven heat distribution within the battery and on the battery pack surface. In thermal imaging, the degree of temperature gradient is represented by the intensity of color changes; the more dramatic the color change, the greater the temperature gradient. Heat flow direction indicates the direction of heat transfer. In thermal imaging, the direction of heat flow can be visually represented by isotherms (lines connecting points of equal temperature). The denser the isotherms, the more obvious the heat flow direction. The overlap of abnormal gas concentration areas indicates the degree of spatial overlap between abnormal gas concentration areas and abnormal temperature areas in thermal imaging. In practical applications, thermal imaging technology and gas sensors can be combined to monitor battery status. Thermal imaging provides temperature distribution information, while gas sensors detect gas concentration. By spatially aligning the data from both and calculating the overlap, a more comprehensive assessment of the battery's safety status can be achieved.

[0119] This application performs dynamic analysis of thermal images to accurately locate battery cells with abnormal temperatures and potential thermal runaway risks. It solves the problems of traditional infrared thermal imaging, such as environmental interference, physical limitations, high installation costs, and insufficient spatiotemporal resolution, providing an efficient and reliable technical solution for the safety monitoring of energy storage battery packs.

[0120] This application eliminates the need for a thermal imager and utilizes existing sensor signals from the BMS (Battery Management System). It addresses the problems of traditional infrared thermal imaging technology, such as being greatly affected by the environment, limited installation space, high cost, and insufficient spatiotemporal resolution of weak signals. It can accurately reflect the thermal state of the battery pack and achieve early abnormal location and precise early warning of thermal runaway.

[0121] In the actual implementation process, the installation position and number of sensors can be adjusted according to the actual battery pack structure and application scenario, appropriate outlier detection and normalization methods can be selected, and the parameters of time series analysis and imaging models can be optimized to improve the accuracy of thermal imaging and the reliability of early warning.

[0122] It should be understood that, although Figure 1-6 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1-6 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0123] This application also provides a system for locating abnormal batteries, such as... Figure 7 As shown, the positioning system includes:

[0124] The first determining module 71 is used to determine the operating data of the battery pack; wherein, the operating data includes surface temperature, internal temperature, gas composition, gas concentration, and casing deformation pressure, and the operating data is obtained based on a sensor array disposed on the battery pack; the sensor array includes a temperature sensor, a gas sensor, a voltage sensor, and a micro-differential pressure sensor; the battery pack includes at least two batteries;

[0125] Processing module 72 is used to perform outlier detection and normalization on the running data to obtain target running data;

[0126] The second determining module 73 is used to perform time series analysis on the target operation data, determine the time characteristics, and obtain the spatial characteristics based on the location information of each battery and the target operation data;

[0127] Image generation module 74 is used to input spatiotemporal features into a thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal features; the spatiotemporal features include the temporal features and the spatial features;

[0128] The third determining module 75 is used to determine the battery anomaly corresponding to the target area in response to the target feature value of the target area in the thermal imaging exceeding a preset threshold; the target feature value includes temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change amplitude, and pressure change rate.

[0129] Optionally, the temperature sensor includes at least one of the following: a flexible thin-film temperature sensor, a distributed optical fiber temperature sensor, a thermocouple sensor, or a thermistor; the flexible thin-film temperature sensor is embedded in the gaps between the batteries in a matrix form, and the distributed optical fiber temperature sensor is arranged in a serpentine pattern around the battery pack.

[0130] Optionally, the outlier detection method includes at least one of the following: Z-score method, IQR method, and isolated forest algorithm;

[0131] The normalization process includes at least one of the following: min-max normalization and Z-score normalization.

[0132] Optionally, the second determining module is also used for:

[0133] The spatiotemporal features are obtained by assigning weights to the temporal and spatial features based on the attention mechanism.

[0134] Optionally, the second determining module is also used for:

[0135] Construct a spatial coordinate matrix based on the physical location of each battery in the battery pack;

[0136] The spatial features are obtained based on the spatial coordinate matrix and the target running data.

[0137] Optionally, the positioning system further includes a calculation module for calculating the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas in the image region corresponding to each battery in the thermal imaging image.

[0138] Based on the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas, the risk level of the battery corresponding to the image area is determined.

[0139] In response to any image region having a risk level exceeding a preset level, a battery anomaly is determined corresponding to that image region.

[0140] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs.

[0141] Figure 8This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the method described in any of the above embodiments. Figure 8 The electronic device 80 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0142] like Figure 8 As shown, the electronic device 80 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 80 may include, but are not limited to: at least one processor 81, at least one memory 82, and a bus 83 connecting different system components (including memory 82 and processor 81).

[0143] Bus 83 includes a data bus, an address bus, and a control bus.

[0144] The memory 82 may include volatile memory, such as random access memory (RAM) 821 and / or cache memory 822, and may further include read-only memory (ROM) 823.

[0145] The memory 82 may also include a program tool 825 (or utility) having a set (at least one) program module 824, such program module 824 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0146] The processor 81 performs various functional applications and data processing, such as the methods provided in any of the above embodiments, by running computer programs stored in the memory 82.

[0147] Electronic device 80 can also communicate with one or more external devices 84 (e.g., keyboard, pointing device, etc.). This communication can be performed via input / output (I / O) interface 85. Furthermore, electronic device 80 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 86. As shown, network adapter 86 communicates with other modules of electronic device 80 via bus 83. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 80, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0148] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0149] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0150] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0151] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0152] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the above embodiments.

[0153] The program code for executing the computer program product of this application can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0154] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0155] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for locating an abnormal battery, characterized in that, The positioning method includes: The operating data of the battery pack is determined; wherein, the operating data includes surface temperature, internal temperature, gas composition, gas concentration, casing deformation pressure, and battery terminal voltage, and the operating data is obtained based on a sensor array disposed on the battery pack; the sensor array includes a temperature sensor, a gas sensor, a voltage sensor, and a micro-differential pressure sensor; the battery pack includes at least two batteries; The running data is subjected to outlier detection and normalization to obtain the target running data; Time series analysis is performed on the target operation data to determine time features, and spatial features are obtained based on the location information of each battery and the target operation data. Attention scores of the time features and spatial features are calculated through linear transformation. The attention scores are normalized to the weights of the time features and spatial features using a predetermined function. The time features and spatial features are then weighted and summed according to the normalized weights to obtain the spatiotemporal features. The spatiotemporal features are input into the thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal features; the spatiotemporal features include the temporal features and the spatial features. In response to the target feature value of the target area in the thermal image exceeding a preset threshold, the battery anomaly corresponding to the target area is determined; the target feature value includes temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change amplitude, and pressure change rate.

2. The positioning method as described in claim 1, characterized in that, The temperature sensor includes at least one of the following: a flexible thin-film temperature sensor, a distributed optical fiber temperature sensor, a thermocouple sensor, or a thermistor; the flexible thin-film temperature sensor is embedded in the gaps between the batteries in a matrix form, and the distributed optical fiber temperature sensor is arranged in a serpentine pattern around the battery pack.

3. The positioning method as described in claim 1, characterized in that, The outlier detection method includes at least one of the following: Z-score method, IQR method, and isolated forest algorithm; The normalization process includes at least one of the following: min-max normalization and Z-score normalization.

4. The positioning method as described in claim 1, characterized in that, The predefined function includes the Softmax function.

5. The positioning method as described in claim 1, characterized in that, The spatial features obtained based on the location information of each battery and the target operation data include: Construct a spatial coordinate matrix based on the physical location of each battery in the battery pack; The spatial features are obtained based on the spatial coordinate matrix and the target running data.

6. The positioning method as described in claim 1, characterized in that, The positioning method further includes: Calculate the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas in the image region corresponding to each battery in the thermal imaging image; Based on the temperature gradient, heat flow direction, and overlap of abnormal gas concentration areas, the risk level of the battery corresponding to the image area is determined. In response to any image region having a risk level exceeding a preset level, a battery anomaly is determined corresponding to that image region.

7. A system for locating an abnormal battery, characterized in that, The positioning system includes: The first determining module is used to determine the operating data of the battery pack; wherein, the operating data includes surface temperature, internal temperature, gas composition, gas concentration, casing deformation pressure, and battery terminal voltage, and the operating data is obtained based on a sensor array disposed on the battery pack; the sensor array includes a temperature sensor, a gas sensor, a voltage sensor, and a micro-differential pressure sensor; the battery pack includes at least two batteries; The processing module is used to perform outlier detection and normalization on the running data to obtain the target running data; The second determining module is used to perform time series analysis on the target operation data to determine the time characteristics, and to obtain the spatial characteristics based on the location information of each battery and the target operation data; An image generation module is used to input spatiotemporal features into a thermal imaging model to generate a thermal image of the battery pack based on the spatiotemporal features; the spatiotemporal features include the temporal features and the spatial features; The third determining module is used to determine the battery anomaly corresponding to the target area in response to the target feature value of the target area in the thermal imaging exceeding a preset threshold; the target feature value includes temperature rise rate, temperature mutation index, gas correlation, voltage anomaly change amplitude, and pressure change rate.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

Images