Method and device for predicting remaining service life of lithium batteries

Abstract

The present invention relates to the technical field of battery life prediction, and specifically to a method and device for predicting the remaining service life of a lithium battery. After obtaining a variety of key parameters of a lithium battery, the solution of the present application uses a weight adjustment algorithm to dynamically adjust the weights based on the historical changes in the parameters and the correlation with the lifespan, so as to make the prediction more accurate and avoid one-sidedness. A matrix that integrates spatiotemporal and battery status characteristics is constructed to integrate multi-dimensional information to provide a comprehensive data basis for prediction, thereby improving accuracy and reliability. According to geographical and environmental parameters, an environmental preset model or a general model is matched to meet the prediction of a specific area and cope with general situations, thereby enhancing the applicability of the method. The environmental preset model is pre-trained based on historical data and can be updated online. It tracks the impact of the environment on battery life in real time, provides timely and accurate predictions, and assists in lithium battery maintenance and replacement decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of battery life prediction, and in particular to a method and device for predicting the remaining service life of a lithium battery. Background Art

[0002] Lithium batteries, due to their high energy density, long cycle life, and low self-discharge, are widely used in electronic devices, electric vehicles, energy storage systems, and many other fields. However, as lithium batteries age, their performance gradually declines, and their remaining service life decreases. Accurately predicting the remaining service life of lithium batteries is crucial for ensuring safe and stable equipment operation, optimizing equipment maintenance plans, and reducing operating costs.

[0003] Existing lithium battery remaining useful life prediction technologies have numerous shortcomings. For one thing, some current methods only consider the battery's inherent state parameters, such as remaining capacity, current and voltage during discharge, and the number of charge and discharge cycles, while ignoring the impact of location and environmental parameters on battery life. Climate conditions and altitude vary across geographic locations, and these environmental factors can significantly impact the performance and lifespan of lithium batteries. For example, high temperatures accelerate the chemical reaction rate within lithium batteries, potentially leading to rapid capacity degradation; low temperatures increase the battery's internal resistance, reducing charge and discharge efficiency. Therefore, failing to consider these factors can significantly reduce the accuracy of prediction results.

[0004] On the other hand, existing prediction methods often fail to dynamically adjust the impact of different parameters on the remaining battery life. Each parameter has a different impact on battery life in different usage scenarios and time periods. Using fixed weights to weigh each parameter would fail to accurately reflect the actual battery state, thus affecting prediction accuracy.

[0005] In summary, the existing lithium battery remaining service life prediction technology has many defects and is difficult to meet the needs of high-precision prediction in practical applications. Summary of the Invention

[0006] To solve the above problems, the present invention discloses a method and device for predicting the remaining service life of a lithium battery.

[0007] To achieve the above objectives, the present application discloses a method for predicting the remaining service life of a lithium battery, comprising the following steps:

[0008] Obtaining key parameters of lithium battery operation, including geographic location parameters, environmental parameters, remaining capacity of the lithium battery, internal battery temperature, current and voltage during discharge, number of charge and discharge cycles, and battery health status parameters;

[0009] monitoring the key parameters and dynamically adjusting the weight of each parameter based on a preset weight adjustment algorithm, wherein the weight adjustment algorithm determines the weight value of each parameter based on the historical change trend of the parameter and the correlation with the remaining service life of the battery;

[0010] Constructing an input feature matrix of the remaining useful life of a lithium battery, wherein the matrix integrates spatiotemporal features and battery state features;

[0011] When the geographic location parameters and / or environmental parameters meet the preset rules, the environmental preset model matching the target climate zone is called to calculate and output the remaining service life of the lithium battery; otherwise, the universal model is used to calculate and output the remaining service life of the lithium battery; wherein, the environmental preset model is a machine learning model pre-trained for the historical degradation data of the target climate zone, and can be updated online to adapt to environmental changes.

[0012] In the above solution, the environmental parameters include ambient temperature, humidity and atmospheric pressure data; the geographical location parameters are obtained by obtaining the latitude and longitude coordinates and altitude values ​​through the GPS module.

[0013] The input feature matrix contains a three-dimensional spatiotemporal tensor associated with a timestamp, wherein the first dimension is the quantized value of the geographic location coordinate, the second dimension is the parameter change rate within the time window, and the third dimension contains the battery state feature vector, which is reduced in dimension using principal component analysis.

[0014] The preset rules are determined using a climate pattern recognition model under a transfer learning framework, specifically including: inputting real-time geographic location parameters into a pre-trained ResNet-Transformer hybrid network to output a climate zone probability distribution; when the target climate zone probability value exceeds a threshold and the environmental parameters match a typical degradation pattern, activating the corresponding environmental preset model.

[0015] On the other hand, the present application also discloses a device for predicting the remaining service life of a lithium battery, the device comprising:

[0016] A parameter acquisition module is used to obtain key parameters of the lithium battery operation, including geographic location parameters, environmental parameters, remaining capacity of the lithium battery, internal battery temperature, current and voltage during discharge, number of charge and discharge cycles, and battery health status parameters;

[0017] a parameter monitoring and weight adjustment module, configured to monitor the key parameters and dynamically adjust the weights of the parameters based on a preset weight adjustment algorithm, wherein the weight adjustment algorithm determines the weight values ​​of the parameters based on the historical change trends of the parameters and their correlation with the remaining battery life;

[0018] A feature matrix construction module is used to construct an input feature matrix of the remaining service life of the lithium battery, wherein the matrix integrates spatiotemporal features and battery state features;

[0019] The remaining service life calculation module calls the environmental preset model that matches the target climate zone to calculate and output the remaining service life of the lithium battery when the geographic location parameters and / or environmental parameters meet the preset rules; otherwise, the universal model is used to calculate and output the remaining service life of the lithium battery; wherein the environmental preset model is a machine learning model pre-trained for the historical degradation data of the target climate zone and can be updated online to adapt to environmental changes.

[0020] The output module is used to display or transmit the calculated remaining service life of the lithium battery.

[0021] In a possible implementation, the device further includes:

[0022] Feature extraction module, used to perform learnable PCA dimensionality reduction on battery operating parameters to generate a spatiotemporal fusion feature matrix;

[0023] The incremental update module performs differential parameter updates when the environment changes suddenly or the prediction error exceeds a threshold.

[0024] The feature extraction module further comprises:

[0025] A hash coding unit, configured to discretize the temperature and the number of cycles into hash values ​​of a preset length;

[0026] The hash value of the preset length is used as the input of the remaining service life calculation module to calculate the remaining service life of the lithium battery.

[0027] The remaining service life calculation module includes a climate judgment module, which is used to match the corresponding climate classification code according to the geographical location coordinates;

[0028] The climate judgment module performs the following operations:

[0029] Mapping latitude and longitude coordinates to climate classification codes;

[0030] Match the preset model identifier through table lookup method.

[0031] The incremental update module triggers the update process when the following conditions are met at the same time:

[0032] Ambient temperature change rate ≥ 5℃ / hour;

[0033] The prediction error exceeds the preset threshold of 10% for three consecutive times;

[0034] Update data transmission uses incremental compression and uses hash checksums to verify data integrity.

[0035] The technical solution in this application, after obtaining multiple key parameters for the operation of the lithium battery, dynamically adjusts the weight of each parameter through a preset weight adjustment algorithm, combined with the historical change trend of the parameters and the correlation with the remaining service life of the battery, so that the prediction of the remaining service life of the lithium battery is more accurate, avoiding the one-sidedness of the prediction under a single fixed weight mode.

[0036] By constructing an input feature matrix that integrates spatiotemporal features and battery status features, the system integrates multi-dimensional information such as time, space, and the battery's state during operation. This multi-feature fusion approach provides a more comprehensive and representative data foundation for predicting the remaining useful life of lithium batteries, thereby improving the accuracy and reliability of the prediction.

[0037] When geographic location parameters and / or environmental parameters meet preset rules, the preset environmental model matching the target climate region is used to calculate and output the remaining service life of the lithium battery; otherwise, a general model is used. This flexible model selection mechanism not only enables accurate predictions for specific climate regions, but also provides effective prediction results in general situations, greatly improving the applicability of this method under different environmental conditions.

[0038] The environmental preset model is a machine learning model pre-trained based on historical degradation data for the target climate region and can be updated online to adapt to environmental changes. This enables the model to track the impact of environmental changes on lithium-ion battery life in real time, ensuring accurate predictions of the remaining useful life of lithium-ion batteries in a constantly changing environment. This enhances the timeliness and reliability of the predictions, providing more timely and accurate decision-making for lithium-ion battery maintenance and replacement. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of a method for predicting the remaining service life of a lithium battery in an embodiment of the present application;

[0040] Figure 2 This is a schematic diagram of the structure of a device for predicting the remaining useful life of a lithium battery in an embodiment of the present application;

[0041] Figure 3 Schematic diagram of the structure of a lithium battery remaining service life prediction device including a feature extraction module and an incremental update module in an embodiment of the present application;

[0042] Figure 4 Schematic diagram of the structure of the feature extraction module and the remaining service life calculation module in the embodiment of the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. The principles and features of the present invention are described below in conjunction with the drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0044] The terms "including" and other similar expressions in the specification or claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions, such as a process, method or system, or device that includes a series of steps or units is not limited to the listed steps or units.

[0045] Example 1: Figure 1 As shown, a method for predicting the remaining service life of a lithium battery includes the following steps:

[0046] S101: Acquire key operating parameters of the lithium battery, including geographic location parameters, environmental parameters, remaining capacity of the lithium battery, internal battery temperature, current and voltage during discharge, number of charge and discharge cycles, and battery health status parameters;

[0047] The purpose of this step is to collect various parameters closely related to the working status of the lithium battery. Among them, the geographical location parameters include the geographical location information of the lithium battery, such as longitude and latitude, which will affect the working environment of the battery, such as the differences in climatic conditions in different regions;

[0048] Environmental parameters include temperature, humidity, air pressure and other environmental factors, which have a significant impact on the performance and life of lithium batteries. For example, high temperature may accelerate the chemical reaction inside the battery and shorten the battery life.

[0049] In this embodiment, the environmental parameters include ambient temperature, humidity and atmospheric pressure data; the geographical location parameters are obtained by the GPS module to obtain the latitude and longitude coordinates and altitude values;

[0050] The remaining capacity of a lithium battery reflects how much power the battery can currently store and is an important indicator for measuring the battery's available energy.

[0051] The actual temperature inside the battery, too high or too low temperature will affect the battery's charging and discharging efficiency and safety;

[0052] Current and voltage during discharge: When the battery is discharging, the changes in current and voltage are monitored in real time. These parameters can reflect the working status and performance of the battery.

[0053] The number of charge and discharge cycles records the number of complete charge and discharge cycles the battery has undergone. Generally speaking, the more cycles there are, the more gradually the battery performance will decline.

[0054] Battery health status parameters are used to evaluate the overall health status of the battery, such as the battery's internal resistance and capacity decay.

[0055] S102: monitoring key parameters and dynamically adjusting the weight of each parameter based on a preset weight adjustment algorithm, where the weight adjustment algorithm determines the weight value of each parameter based on the historical change trend of the parameter and its correlation with the remaining battery life;

[0056] This step continuously monitors the key parameters obtained in S101, and then the weight adjustment algorithm determines the weight value of each parameter based on the historical change trend of the parameter and the correlation with the remaining battery life. For example, if it is found that the change trend of a certain environmental parameter over a period of time in the past is strongly correlated with the shortening of battery life, then the weight of the parameter will be increased accordingly; conversely, if a certain parameter has little to do with battery life, its weight may be reduced. This dynamic adjustment of weights can make subsequent predictions of the remaining battery life more accurate, because different parameters have different effects on battery life under different circumstances.

[0057] S103: Constructing an input feature matrix of the remaining service life of the lithium battery, where the matrix integrates spatiotemporal features and battery status features;

[0058] This step integrates the various parameters obtained previously to construct an input feature matrix. This matrix incorporates spatiotemporal features (such as spatial features represented by geographic location parameters and time-related parameter changes) and battery status features (such as remaining battery capacity and internal temperature). This approach comprehensively represents multiple aspects of the battery's remaining lifespan, providing more comprehensive and accurate input data for subsequent model calculations.

[0059] S104: When the geographic location parameters and / or environmental parameters meet the preset rules, call the environmental preset model that matches the target climate zone to calculate and output the remaining service life of the lithium battery; otherwise, use the general model to calculate and output the remaining service life of the lithium battery; wherein, the environmental preset model is a machine learning model pre-trained for the historical degradation data of the target climate zone, and can be updated online to adapt to environmental changes.

[0060] Preset rules might mean that when a battery's geographic location or environmental conditions meet a specific range or standard, the current environment is considered unique and requires prediction using a machine learning model pre-trained on historical degradation data specifically for that target climate region. For example, battery performance in cold polar regions and hot deserts may differ from that in general environments, necessitating the use of a specially trained model to more accurately predict remaining life.

[0061] Otherwise, when the preset rules are not met, the general model is used to calculate and output the remaining service life of the lithium battery. The general model is a model applicable to general environmental conditions.

[0062] Furthermore, the environmental preset model can be updated online to adapt to environmental changes. As environmental conditions change over time or new data is generated, online model updates ensure that the model remains adaptable to the current environment, improving the accuracy and reliability of predictions.

[0063] Exemplary, environmental preset models: pre-trained models for typical climates such as tropical / cold / plateau

[0064] Universal model: a benchmark model trained on global data

[0065] Switching logic: Determine the current climate zone through a decision tree or rule engine (e.g., temperature > 40°C and humidity > 80% triggers the tropical mode)

[0066] In one possible embodiment, the input feature matrix includes a three-dimensional spatiotemporal tensor associated with a timestamp, where the first dimension is the quantized value of the geographic location coordinate, the second dimension is the parameter change rate within the time window, and the third dimension contains the battery state feature vector, and the feature vector is reduced in dimensionality through principal component analysis.

[0067] In this embodiment, the input feature matrix is ​​the key data structure of the model for predicting the remaining life of lithium batteries. It exists in the form of a three-dimensional space-time tensor and is associated with a timestamp. It can reflect the time sequence of the data and is conducive to analyzing the changes in battery performance over time.

[0068] The first dimension is the quantitative value of the geographic location coordinates: the location of the lithium battery affects its performance and lifespan. The geographic location coordinates such as longitude and latitude are quantified into numerical values ​​according to rules. The model can then consider the impact of different regional environments (such as climate differences) on the battery based on this value.

[0069] The second dimension is the parameter change rate within a time window: Key parameters are monitored within a specific time period (time window) and their rate of change, such as the battery temperature change rate, is calculated. This rate of change reflects the dynamic trend of the parameter and provides useful information for predicting battery life.

[0070] The third dimension is the battery status feature vector, reduced in dimension using principal component analysis. The battery status feature vector combines multiple parameters, such as remaining capacity, temperature, current, and voltage. These parameters are inherently high-dimensional and potentially redundant. Using principal component analysis (PCA) for dimensionality reduction preserves key information, removes redundancy, and reduces data dimensionality, improving model computational efficiency and performance, enabling better prediction of remaining battery life.

[0071] In one possible implementation, the preset rules are determined using a climate pattern recognition model under a transfer learning framework, specifically including: inputting real-time geographic location parameters into a pre-trained ResNet-Transformer hybrid network to output a climate zone probability distribution; when the target climate zone probability value exceeds a threshold and the environmental parameters match a typical degradation pattern, activating the corresponding environmental preset model.

[0072] Transfer learning refers to leveraging knowledge already learned from other related tasks or data (pre-trained models) and applying it to the current task to improve the model's learning efficiency and performance. The role of the climate pattern recognition model is to identify the climate pattern in which the lithium battery is exposed, thus providing a basis for subsequent decision-making.

[0073] The lithium battery's real-time geographic location parameters (such as longitude and latitude) are input into a pre-trained hybrid network. After computational processing, the network outputs a probability distribution for different climate zones. For each possible climate zone (such as tropical, temperate, or frigid), the network generates a probability distribution for the current lithium battery location. For example, the output might indicate a probability of 0.6 for the tropical zone, 0.3 for the temperate zone, and 0.1 for the frigid zone.

[0074] For example, assume a probability threshold is set. When the calculated probability of a target climate zone (i.e., the specific climate zone of interest) exceeds this threshold, it indicates that the current lithium battery is likely to be in that target climate zone. For example, if the threshold is set to 0.5, and the probability of a target climate zone is 0.6, this condition is met.

[0075] In addition to considering the probability of the climate zone corresponding to the geographic location, it is also necessary to check whether the current environmental parameters (such as temperature, humidity, and air pressure) match the typical degradation pattern of the target climate zone. Typical degradation patterns refer to the common performance degradation characteristics and patterns exhibited by lithium batteries during long-term use in that climate zone. For example, in climates with high temperature and high humidity, lithium batteries may exhibit specific patterns of capacity decay and internal resistance increase. Only when the environmental parameters match the typical degradation pattern are the conditions of the preset rules considered to be met.

[0076] When both of these conditions are met—that is, the probability value for the target climate region exceeds the threshold and the environmental parameters match a typical degradation pattern—the preset environmental model for that target climate region is activated. This preset environmental model is a machine learning model pre-trained based on historical degradation data for that target climate region. Once activated, it can be used to more accurately calculate the remaining useful life of lithium batteries.

[0077] Example 2: Figure 2 As shown, a device for predicting the remaining service life of a lithium battery comprises:

[0078] The parameter acquisition module is used to obtain the key parameters of the lithium battery operation, including geographical location parameters, environmental parameters, remaining capacity of the lithium battery, internal battery temperature, current and voltage during discharge, number of charge and discharge cycles, and battery health status parameters;

[0079] After the parameter acquisition module obtains these key parameters, its output is passed to the parameter monitoring and weight adjustment module and the feature matrix construction module;

[0080] The parameter monitoring and weight adjustment module monitors key parameters and dynamically adjusts their weights based on a preset weight adjustment algorithm. This algorithm determines the weights of each parameter based on the parameter's historical trend and its correlation with the remaining battery life. It receives key parameter data from the parameter acquisition module for monitoring and weight adjustment calculations. After completing the weight adjustment, it is not directly connected to other modules, but the parameter weights it calculates will affect the feature matrix construction module, as the construction of the feature matrix requires comprehensive consideration of all parameters and their weights.

[0081] Feature matrix construction module, used to construct the input feature matrix of the remaining service life of lithium batteries. The matrix integrates spatiotemporal features and battery status features;

[0082] The parameter acquisition module obtains key parameter data and, referring to the parameter weights calculated by the parameter monitoring and weight adjustment module, integrates this information to construct an input feature matrix. This constructed feature matrix is ​​then passed to the remaining service life calculation module as the input basis for calculating the remaining battery life.

[0083] The remaining service life calculation module calls the environmental preset model that matches the target climate zone to calculate and output the remaining service life of the lithium battery when the geographic location parameters and / or environmental parameters meet the preset rules; otherwise, the universal model is used to calculate and output the remaining service life of the lithium battery; wherein, the environmental preset model is a machine learning model pre-trained for the historical degradation data of the target climate zone, and can be updated online to adapt to environmental changes.

[0084] The output module is used to display or transmit the calculated remaining service life of the lithium battery.

[0085] like Figure 3 As shown, in a possible implementation manner, the device further includes:

[0086] The feature extraction module performs learnable principal component analysis (PCA) dimensionality reduction on battery operating parameters to generate a spatiotemporal fusion feature matrix. The feature extraction module processes battery operating parameters, specifically using a learnable principal component analysis (PCA) dimensionality reduction method to generate a spatiotemporal fusion feature matrix. After performing learnable PCA dimensionality reduction on the battery operating parameters, the resulting spatiotemporal fusion feature matrix combines feature information from both temporal and spatial dimensions. For example, geographic location parameters represent spatial characteristics, while the temporal changes in battery operating parameters (such as temperature and capacity changes at different time points) represent temporal characteristics. This fusion integrates multiple aspects of information into a single matrix, providing a more concise and representative data format for subsequent analysis and prediction.

[0087] The input data for this module comes from the battery operating parameters acquired by the parameter acquisition module. The resulting spatiotemporal fusion feature matrix provides more refined feature data for the feature matrix construction module, assisting it in constructing a more effective input feature matrix. Furthermore, it can directly contribute to the remaining useful life calculation module, providing higher-quality input for battery remaining life prediction.

[0088] The incremental update module performs differential parameter updates when the environment suddenly changes or the prediction error exceeds a threshold. The incremental update module requires environmental parameter information (from the parameter acquisition module) to determine whether a sudden change has occurred. It also requires prediction error information from the remaining service life calculation module to determine whether the threshold has been exceeded. When performing a differential parameter update, it updates parameters in related modules, such as the environmental preset model and the general model, affecting subsequent calculations and predictions in these modules.

[0089] For example, the incremental update module triggers the update process when the following conditions are met simultaneously:

[0090] Ambient temperature change rate ≥ 5℃ / hour;

[0091] The prediction error exceeds the preset threshold of 10% for three consecutive times;

[0092] Update data transmission uses incremental compression, and data integrity is verified using a hash checksum. By dually evaluating the temperature mutation rate and error persistence, false triggering can be avoided. This incremental compression approach significantly reduces system resource consumption while ensuring update reliability, making it ideal for edge computing deployments.

[0093] like Figure 4 As shown, the feature extraction module further includes:

[0094] A hash coding unit, which is used to discretize the temperature and the number of cycles into hash values ​​of a preset length;

[0095] Coding example:

[0096] Temperature range | Hash code

[0097] [-20,0)℃|0x5A

[0098] [0,25)℃|0x3B

[0099] [25,60]℃|0x8F

[0100] Replace floating point operations,

[0101] The hash value of the preset length is used as the input of the remaining service life calculation module to calculate the remaining service life of the lithium battery.

[0102] The hash encoding unit discretizes specific battery operating parameters, namely temperature and cycle count, and converts them into hash values ​​of a preset length. Discretization divides a continuous range of values ​​into discrete intervals, simplifying data representation and processing. Hash encoding maps the originally continuous temperature and cycle count values ​​to specific hash values ​​with a fixed, preset length, facilitating subsequent data processing and calculations.

[0103] Typically, parameters such as temperature and cycle count are represented as floating-point numbers during data processing. Floating-point operations are relatively complex and, in some scenarios, can consume significant computational resources and time. By discretizing temperature and cycle count into hash values, these hash values ​​are used instead of the original floating-point numbers for subsequent operations and processing. Hash values ​​are typically fixed-length codes, making them more efficient and concise in storage and computation than floating-point numbers. This reduces computational effort and improves data processing speed and efficiency.

[0104] The hash value of a preset length obtained after processing by the hash encoding unit is passed as input to the remaining service life calculation module. The remaining service life calculation module uses the hash value and other relevant input information (such as the characteristics obtained from the processing of other battery operating parameters) to calculate the remaining service life of the lithium battery.

[0105] The remaining useful life calculation module includes a climate judgment module, which is used to match the corresponding climate classification code according to the geographical location coordinates;

[0106] The climate judgment module performs the following operations:

[0107] Mapping latitude and longitude coordinates to climate classification codes;

[0108] Match the preset model identifier through table lookup method.

[0109] For example, to map longitude and latitude coordinates to 3-digit climate classification codes, the coding rules include:

[0110] First: Temperature zone (1-tropical, 2-subtropical, etc.)

[0111] Second: Humidity characteristics (A-dry, B-humid...)

[0112] Last place: Seasonal pattern (S - rainy summer, W - rainy winter)

[0113] Finally, the preset model identifier is matched through table lookup method.

[0114] The climate determination module determines the corresponding climate category based on the lithium battery's geographic location and finds a matching preset model identifier. The remaining service life calculation module uses the preset model identifier provided by the climate determination module to call the corresponding model and combine it with other input information (such as battery operating parameters) to accurately calculate the remaining service life of the lithium battery.

[0115] The technical means disclosed in the solutions of the present invention are not limited to those disclosed in the above-mentioned embodiments, but also include technical solutions composed of any combination of the above-mentioned technical features. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for predicting the remaining service life of a lithium battery, characterized in that: The following steps are involved: Obtaining key parameters of lithium battery operation, including geographic location parameters, environmental parameters, remaining capacity of the lithium battery, internal battery temperature, current and voltage during discharge, number of charge and discharge cycles, and battery health status parameters; monitoring the key parameters and dynamically adjusting the weight of each parameter based on a preset weight adjustment algorithm, wherein the weight adjustment algorithm determines the weight value of each parameter based on the historical change trend of the parameter and the correlation with the remaining service life of the battery; Constructing an input feature matrix of the remaining useful life of a lithium battery, wherein the matrix integrates spatiotemporal features and battery state features; When the geographic location parameters and / or environmental parameters meet the preset rules, the environmental preset model matching the target climate zone is called to calculate and output the remaining service life of the lithium battery; otherwise, the universal model is used to calculate and output the remaining service life of the lithium battery; wherein, the environmental preset model is a machine learning model pre-trained for the historical degradation data of the target climate zone, and can be updated online to adapt to environmental changes.

2. The method for predicting the remaining service life of a lithium battery according to claim 1, wherein: The environmental parameters include ambient temperature, humidity and atmospheric pressure data; the geographical location parameters are obtained by the GPS module to obtain the latitude and longitude coordinates and altitude values.

3. The method for predicting the remaining service life of a lithium battery according to claim 1, wherein: The input feature matrix contains a three-dimensional spatiotemporal tensor associated with a timestamp, wherein the first dimension is the quantized value of the geographic location coordinate, the second dimension is the parameter change rate within the time window, and the third dimension contains the battery state feature vector, which is reduced in dimension using principal component analysis.

4. The method for predicting the remaining service life of a lithium battery according to claim 1, wherein: The preset rules are determined using a climate pattern recognition model under a transfer learning framework, specifically including: inputting real-time geographic location parameters into a pre-trained ResNet-Transformer hybrid network to output a climate zone probability distribution; when the target climate zone probability value exceeds a threshold and the environmental parameters match a typical degradation pattern, activating the corresponding environmental preset model.

5. A device for predicting the remaining service life of a lithium battery, characterized in that: include: A parameter acquisition module is used to obtain key parameters of the lithium battery operation, including geographic location parameters, environmental parameters, remaining capacity of the lithium battery, internal battery temperature, current and voltage during discharge, number of charge and discharge cycles, and battery health status parameters; a parameter monitoring and weight adjustment module, configured to monitor the key parameters and dynamically adjust the weights of the parameters based on a preset weight adjustment algorithm, wherein the weight adjustment algorithm determines the weight values ​​of the parameters based on the historical change trends of the parameters and their correlation with the remaining battery life; A feature matrix construction module is used to construct an input feature matrix of the remaining service life of the lithium battery, wherein the matrix integrates spatiotemporal features and battery state features; A remaining service life calculation module, which, when the geographic location parameters and / or environmental parameters meet preset rules, calls a preset environmental model that matches the target climate zone to calculate and output the remaining service life of the lithium battery; otherwise, uses a universal model to calculate and output the remaining service life of the lithium battery; wherein the preset environmental model is a machine learning model pre-trained based on historical degradation data for the target climate zone and can be updated online to adapt to environmental changes; The output module is used to display or transmit the calculated remaining service life of the lithium battery.

6. The device for predicting the remaining service life of a lithium battery according to claim 5, characterized in that: The device further comprises: Feature extraction module, used to perform learnable PCA dimensionality reduction on battery operating parameters to generate a spatiotemporal fusion feature matrix; The incremental update module performs differential parameter updates when the environment changes suddenly or the prediction error exceeds the threshold, and verifies the integrity of the updated data.

7. The device for predicting the remaining service life of a lithium battery according to claim 6, characterized in that: The feature extraction module further comprises: A hash coding unit, configured to discretize the temperature and the number of cycles into hash values ​​of a preset length; The hash value of the preset length is used as the input of the remaining service life calculation module to calculate the remaining service life of the lithium battery.

8. The device for predicting the remaining service life of a lithium battery according to claim 6, characterized in that: The remaining service life calculation module includes a climate judgment module, which is used to match the corresponding climate classification code according to the geographical location coordinates; The climate judgment module performs the following operations: Mapping latitude and longitude coordinates to climate classification codes; Match the preset model identifier through table lookup method.

9. The device for predicting the remaining service life of a lithium battery according to claim 6, characterized in that: The incremental update module triggers the update process when the following conditions are met at the same time: Ambient temperature change rate ≥ 5℃ / hour; The prediction error exceeds the preset threshold of 10% for three consecutive times; The incremental update module adopts an incremental compression method when transmitting updated data, and performs integrity verification of the updated data through a hash check code.

Images