Liquid cooling system battery life prediction method, device, electronic device and storage medium

Abstract

The present invention discloses a method, device, electronic device and storage medium for predicting the life of a battery in a liquid cooling system, and relates to the technical field related to battery life prediction. The method includes: partitioning the battery modules based on the structural attribute information of the target liquid cooling system battery; sequentially deploying a multi-sensor sensor network to collect and obtain the working data streams of M partitioned battery modules; building a battery module life prediction channel; matching and mapping the working data streams of the M partitioned battery modules to the battery module life prediction channel for activation prediction analysis; performing correlation impact analysis on the M battery module life prediction information according to the battery module distribution characteristic information to determine the liquid cooling system battery life prediction result. The method solves the technical problems of insufficient battery life prediction accuracy and inability to effectively reflect the non-uniform aging characteristics of liquid cooling system batteries in the prior art, and achieves the technical effects of optimizing battery health management strategies, improving the accuracy of liquid cooling system battery life prediction and the reliability of the liquid cooling system.

Description

Technical Field

[0001] The present application relates to the technical field related to battery life prediction, and specifically to a method, device, electronic device and storage medium for predicting battery life in a liquid cooling system. Background Art

[0002] Lithium-ion batteries are key components of electric vehicles, large-scale energy storage systems, and various portable electronic devices. However, battery performance gradually degrades during actual operation, affecting the reliability and safety of the overall system. Accurately predicting battery life is key to optimizing battery management and extending system service life. Traditional battery life prediction relies on accelerated aging experiments and simulation analysis in laboratory environments. This makes it difficult to fully reflect the battery degradation characteristics under complex actual operating conditions. This is especially true in liquid-cooled battery systems, where uneven coolant distribution, temperature gradients within battery modules, and current load differences in different regions lead to significant differences in the degradation rates of different battery components, resulting in localized over-discharge or over-charge, which in turn affects the performance and safety of the overall system. Furthermore, existing battery life predictions ignore the spatial heterogeneity of battery systems, such as the impact of factors such as heat dissipation efficiency and uneven current distribution in different partitions of the liquid-cooled system on battery aging, resulting in limited prediction accuracy.

[0003] Therefore, in the current related technologies, there are technical problems such as insufficient accuracy in battery life prediction and inability to effectively reflect the non-uniform aging characteristics of liquid cooling system batteries. Summary of the Invention

[0004] This application solves the technical problems in the prior art of insufficient battery life prediction accuracy and inability to effectively reflect the non-uniform aging characteristics of liquid cooling system batteries by providing a liquid cooling system battery life prediction method, device, electronic device and storage medium, thereby achieving the technical effect of optimizing battery health management strategies, improving the accuracy of liquid cooling system battery life prediction and the reliability of the liquid cooling system.

[0005] The present application provides a liquid cooling system battery life prediction method, which includes: partitioning the battery modules based on the structural attribute information of the target liquid cooling system battery to obtain M liquid cooling system partitioned battery modules; sequentially deploying a multi-sensor sensor network on the M liquid cooling system partitioned battery modules, and collecting and acquiring the working data streams of the M partitioned battery modules through the multi-sensor network; building a battery module life prediction channel, wherein the battery module life prediction channel includes a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; matching and mapping the M partitioned battery module working data streams to the battery module life prediction channel for activation prediction analysis to obtain M battery module life prediction information; performing correlation impact analysis on the M battery module life prediction information according to the distribution characteristic information of the M liquid cooling system partitioned battery modules to determine the liquid cooling system battery life prediction result.

[0006] In a possible implementation, the liquid cooling system battery life prediction method also performs the following processing: obtaining a battery attribute factor set, the battery attribute factor set including module layout, arrangement, spatial connection relationship, battery cell size and shape, thermal physical parameters and liquid cooling integration form; extracting factor parameters of the structural attribute information of the target liquid cooling system battery according to the battery attribute factor set to obtain a liquid cooling battery factor parameter set; constructing a liquid cooling battery partitioning strategy, the liquid cooling battery partitioning strategy including battery spatial position adjacency rules, partition battery quantity balance rules and partition battery liquid cooling balance rules; partitioning the liquid cooling battery factor parameter set into battery modules based on the liquid cooling battery partitioning strategy to obtain M liquid cooling system partitioned battery modules.

[0007] In a possible implementation, the liquid-cooling system battery life prediction method further performs the following processing: prioritizing each partitioning rule in the liquid-cooling battery partitioning strategy to determine a liquid-cooling battery partitioning rule sequence; generating a liquid-cooling battery distribution topology map based on the liquid-cooling battery partitioning strategy according to the liquid-cooling battery partitioning rule sequence to obtain a plurality of initial partitioned battery modules; performing simulation verification testing and partition iterative optimization on the plurality of initial partitioned battery modules to obtain the M liquid-cooling system partitioned battery modules.

[0008] In a possible implementation, the liquid cooling system battery life prediction method also performs the following processing: collecting and obtaining a liquid cooling system battery working life data set, the liquid cooling system battery working life data set includes historical liquid cooling battery structure attribute data, battery module working data and corresponding battery life data; clustering analysis is performed on the historical liquid cooling battery structure attribute data according to a preset number of clusters to obtain conventional battery clustering clusters and personalized battery clustering clusters, wherein the preset number of clusters is 2; based on the conventional battery clustering clusters and personalized battery clustering clusters, the battery module working data and the corresponding battery life data are classified and integrated to obtain a conventional battery life data set and a personalized battery life data set; life prediction training is performed on the conventional battery life data set and the personalized battery life data set to build a battery module life prediction channel.

[0009] In a possible implementation, the liquid cooling system battery life prediction method further performs the following processing: feature identification is performed on the conventional battery life data set and the personalized battery life data set respectively to obtain a conventional battery life sample set and a personalized battery life sample set; a deep neural network structure is used to perform prediction training and iterative optimization on the conventional battery life sample set and the personalized battery life sample set to generate a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; the conventional battery life prediction branch channel and the personalized battery life prediction branch channel are connected in parallel to fuse the battery module life prediction channel.

[0010] In a possible implementation, the liquid cooling system battery life prediction method also performs the following processing: calculating the mean battery module life information of the M battery module life prediction information; screening and obtaining N battery module life information that does not reach the mean battery module life information from the M battery module life prediction information; performing correlation impact analysis on the N battery module life information according to the distribution characteristic information of the M liquid cooling system partition battery modules to obtain N battery life impact factors; performing impact correction on the mean battery module life information based on the N battery life impact factors to determine the liquid cooling system battery life prediction result.

[0011] In a possible implementation, the liquid cooling system battery life prediction method further performs the following processing: performing a correlation impact analysis of a preset spatial radiation distance on each battery module in the N battery module life information in accordance with the distribution characteristic information of the M liquid cooling system partitioned battery modules, and determining the life attenuation coefficients of the N battery modules; obtaining the radiation impact quantity of the N battery modules in the life information of the N battery modules; and performing a weighted calculation on the life attenuation coefficients of the N battery modules based on the radiation impact quantity of the N battery modules to obtain the N battery life impact factors.

[0012] The present application also provides a liquid cooling system battery life prediction device, which includes: a battery module partitioning unit, which is used to partition the battery modules based on the structural attribute information of the target liquid cooling system battery to obtain M liquid cooling system partitioned battery modules; a working data stream acquisition unit, which is used to sequentially deploy a multi-sensor sensor network on the M liquid cooling system partitioned battery modules, and collect and acquire the working data streams of the M partitioned battery modules through the multi-sensor network; a life prediction channel construction unit, which is used to construct a battery module life prediction channel, and the battery module life prediction channel includes a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; an activation prediction analysis unit, which is used to match and map the M partitioned battery module working data streams to the battery module life prediction channel to perform activation prediction analysis to obtain M battery module life prediction information; and a correlation impact analysis unit, which is used to perform correlation impact analysis on the M battery module life prediction information according to the distribution characteristic information of the M liquid cooling system partitioned battery modules to determine the liquid cooling system battery life prediction result.

[0013] The present application also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing a method for predicting the battery life of a liquid cooling system when executing the executable instructions stored in the memory.

[0014] The present application also provides a computer-readable storage medium, comprising: a computer program stored thereon, which implements a method for predicting the battery life of a liquid cooling system when the program is executed by a processor.

[0015] The liquid cooling system battery life prediction method, device, electronic device and storage medium proposed in this application are intended to partition the battery modules based on the structural attribute information of the target liquid cooling system battery; sequentially deploy a multi-sensor sensor network to collect and obtain the working data streams of the M partitioned battery modules; establish a battery module life prediction channel; match and map the working data streams of the M partitioned battery modules to the battery module life prediction channel for activation prediction analysis; perform correlation impact analysis on the M battery module life prediction information according to the battery module distribution characteristic information to determine the liquid cooling system battery life prediction result. This solves the technical problems of insufficient battery life prediction accuracy and inability to effectively reflect the non-uniform aging characteristics of liquid cooling system batteries in the prior art, and achieves the technical effect of optimizing battery health management strategies, improving the accuracy of liquid cooling system battery life prediction and the reliability of the liquid cooling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A flow chart of a method for predicting battery life in a liquid cooling system according to an embodiment of the present application.

[0018] Figure 2 This is a schematic diagram of the structure of the liquid cooling system battery life prediction device provided in an embodiment of the present application.

[0019] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0020] Explanation of the accompanying symbols: battery module partitioning unit 10, working data stream acquisition unit 20, life prediction channel building unit 30, activation prediction analysis unit 40, correlation impact analysis unit 50, input device 401, processor 402, memory 403, output device 404. DETAILED DESCRIPTION

[0021] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0022] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0023] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0024] The present application embodiment provides a method for predicting the battery life of a liquid cooling system, such as Figure 1 As shown, the method includes:

[0025] Step S100 , partitioning the battery modules based on the structural attribute information of the target liquid cooling system battery to obtain M liquid cooling system partitioned battery modules.

[0026] Preferably, in the target liquid-cooled battery system, the physical layout, coolant flow path, thermal distribution characteristics and electrical connection method of the battery module are different, and the battery cells in different areas may exhibit different working conditions and aging trends. Then, the battery module is reasonably partitioned according to the structural property information of the liquid cooling system to obtain M liquid-cooling system partitioned battery modules, where M is a positive integer greater than 1, representing the total number of partitions of the battery module. Specifically, the structural property information of the target liquid-cooled system battery refers to the physical and electrical characteristics that affect key factors such as the internal temperature, current distribution, and cooling efficiency of the battery module, which may include battery arrangement methods, such as series / parallel structure, module stacking method; liquid cooling pipe layout, such as coolant flow direction, flow channel design, inlet and outlet positions; thermal management characteristics, such as heat dissipation efficiency and temperature gradient distribution in different areas; electrical connection methods, such as tab connection methods and current path distribution; mechanical fixing structures, such as the influence of battery arrangement spacing and fixing brackets on heat dissipation.

[0027] Preferably, battery modules are partitioned according to the structural property information of the target liquid-cooled system battery, which may include thermal distribution partitioning, cooling efficiency partitioning, current distribution partitioning, and mechanical structure partitioning. Thermal distribution partitioning refers to dividing battery cells with similar temperatures into the same partition based on temperature sensor data or computational fluid dynamics simulation; cooling efficiency partitioning refers to dividing areas with faster and slower coolant flow into high-efficiency cooling areas and low-efficiency cooling areas based on the design of the liquid-cooling channel; current distribution partitioning refers to dividing areas with higher and lower current densities based on the electrical connection method of the battery module; mechanical structure partitioning refers to partitioning by physical modules, such as by battery pack module, to obtain upper modules, lower modules, edge modules, etc. Ultimately, the entire target liquid-cooled battery system is divided into M battery module partitions, each with similar thermal, electrical, and mechanical characteristics. For example, partition 1 is close to the coolant inlet, with better heat dissipation, lower temperature, and slower aging; partition 2 is located at the end of the liquid-cooling channel, with lower cooling efficiency, higher temperature, and faster aging; partition 3 is a current-concentrated area with greater internal resistance heating and more obvious cycle attenuation. Different partitions have different battery aging patterns, so data can be collected and battery life predictions can be made in a targeted manner to facilitate more accurate monitoring and prediction of battery life.

[0028] Furthermore, step S100 also includes step S110, obtaining a battery attribute factor set, wherein the battery attribute factor set includes module layout, arrangement, spatial connection relationship, battery cell size and shape, thermal physical parameters and liquid cooling integration form; step S120, extracting factor parameters of the structural attribute information of the target liquid cooling system battery according to the battery attribute factor set to obtain a liquid cooling battery factor parameter set; step S130, constructing a liquid cooling battery partitioning strategy, wherein the liquid cooling battery partitioning strategy includes a battery spatial position adjacency rule, a partition battery quantity balance rule and a partition battery liquid cooling balance rule; step S140, partitioning the liquid cooling battery factor parameter set into battery modules based on the liquid cooling battery partitioning strategy to obtain M liquid cooling system partitioned battery modules.

[0029] Preferably, by performing industrial CT scanning, infrared thermal imaging scanning, three-dimensional laser measurement, computational fluid dynamics simulation on the target liquid cooling system battery and combining it with battery design data, a set of key characteristic parameters of the thermal-electrical-mechanical properties of the liquid cooling battery system, that is, a set of battery property factors, mainly including module layout, arrangement, spatial connection relationship, cell size and shape, thermal physical parameters and liquid cooling integration form, wherein the module layout refers to the spatial arrangement of modules in the battery pack, such as matrix, stacked, and special-shaped arrangement; the arrangement refers to the cell-level arrangement, such as parallel / series topology, tab orientation; the spatial connection relationship refers to the mechanical fixation method between modules, such as bolt connection, welding, and electrical connection path; the cell size and shape refers to the physical size of the single cell, such as cylindrical / square / soft pack; the thermal physical parameters refer to the material properties of the cell such as specific heat capacity, thermal conductivity, contact thermal resistance, etc.; the liquid cooling integration form includes the cooling plate structure, such as serpentine / parallel flow channels, cooling medium and interface distribution.

[0030] Preferably, factor parameters are extracted from the structural attribute information of the target liquid cooling system battery according to the battery attribute factor set. Specifically, spatial data such as the module three-dimensional coordinates and cell spacing are obtained through CAD model analysis, an equivalent thermal resistance network model is constructed based on the thermophysical parameters, an impedance distribution matrix is ​​established according to the connection relationship, and fluid parameters such as the coolant flow rate and pressure drop in each area are calculated to obtain a liquid cooling battery factor parameter set, which may include spatial parameters such as the module center coordinates (x, y, z), cell spacing, thermal parameters such as local thermal conductivity, contact thermal resistance, electrical parameters such as loop equivalent impedance, current distribution weight, liquid cooling parameters, flow velocity of the flow channel section, and inlet and outlet pressure difference.

[0031] Preferably, a liquid-cooled battery partitioning strategy is constructed, including a battery spatial position adjacency rule, a partition battery quantity balance rule, and a partition battery liquid cooling balance rule, wherein the battery spatial position adjacency rule ensures that the battery modules in the same partition are closely adjacent in physical space, avoiding thermal management failure caused by cross-regional partitioning, and based on three-dimensional coordinate data, spatial clustering such as K-means or hierarchical clustering is used to divide adjacent modules, and a maximum span limit is set, such as not more than 2 times the battery cell diameter; the partition battery quantity balance rule ensures that the number of batteries in each partition is relatively balanced, avoiding overload of some partitions and insufficient utilization of other partitions. Under the premise of satisfying adjacency, the partition boundary is adjusted through a greedy algorithm so that the number of batteries in each partition falls within a preset range, such as a deviation of ±5%; the partition battery liquid cooling balance rule ensures that the cooling efficiency of each partition is similar, avoiding local overheating or overcooling, and based on CFD simulation data, the coolant flow rate, temperature distribution and other parameters are balanced and optimized, with the goal of making the coolant flow difference in each partition <10%.

[0032] Preferably, the liquid-cooled battery factor parameter set is partitioned into battery modules based on the liquid-cooled battery partitioning strategy. Specifically, the liquid-cooled battery factor parameter set is converted into a multidimensional feature space, and then the DBSCAN algorithm is used for adaptive clustering, including setting the ε neighborhood radius to include thermal / electrical coupling effects, and the minimum number of samples is dynamically calculated according to the scale of the liquid cooling unit to obtain M liquid-cooled system partitioned battery modules; then the partitioning scheme is verified by a constraint satisfaction problem (CSP) solver to ensure that the battery module partitioning meets adjacency, quantity balance and optimized liquid cooling uniformity.

[0033] Furthermore, step S140 also includes step S141, prioritizing each partitioning rule in the liquid-cooled battery partitioning strategy to determine a liquid-cooled battery partitioning rule sequence; step S142, generating a liquid-cooled battery distribution topology map based on the liquid-cooled battery factor parameter set; step S143, partitioning the liquid-cooled battery distribution topology map into battery modules based on the liquid-cooled battery partitioning strategy according to the liquid-cooled battery partitioning rule sequence to obtain a plurality of initial partitioned battery modules; step S144, performing simulation verification testing and partition iterative optimization on the plurality of initial partitioned battery modules to obtain the M liquid-cooling system partitioned battery modules.

[0034] Preferably, according to the safety and performance requirements of the liquid cooling system, the partitioning rules in the liquid-cooled battery partitioning strategy are prioritized, the battery spatial position adjacency rule is set to the highest priority, the partition battery quantity balance rule is set to the medium priority, and the partition battery liquid cooling balance rule is set to the low priority, that is, the liquid-cooled battery partitioning rule sequence is determined. If the liquid-cooled battery partitioning strategy cannot be satisfied at the same time, the spatial adjacency is prioritized, and then the quantity and liquid cooling balance are adjusted through iterative optimization; then, based on the liquid-cooled battery factor parameter set, a topology map reflecting the multi-physical field characteristics of the battery system is constructed. Specifically, the geometric layout of the battery module partitions is converted into three-dimensional point cloud data, edge relationships are added according to electrical connections and cooling paths, and then the multi-dimensional parameters of thermal, electrical, and fluid mechanics are integrated to generate a liquid-cooled battery distribution topology map, wherein the node is each battery module with accompanying attributes such as coordinates, thermal resistance, current load, etc., and the edge is the connection relationship between modules, such as physical adjacency, electrical coupling, thermal interaction, and heat conduction efficiency, current density, or coolant flow influence is expressed as edge weight.

[0035] Preferably, the liquid-cooled battery distribution topology is partitioned into battery modules according to the liquid-cooled battery partitioning rule sequence and based on the liquid-cooled battery partitioning strategy, that is, the liquid-cooled battery partitioning rules are applied in sequence according to the battery partitioning rule priority. Specifically, a spatial clustering algorithm is used to generate physically adjacent initial partitions, and the partition boundaries are adjusted to meet the quantity balance requirements. Then, the partitions whose liquid cooling parameters do not meet the standards are fine-tuned, such as merging or splitting, and then multiple initial partitioned battery modules are output, which may not fully meet all battery partitioning rules. Among them, if the liquid cooling balance conflicts with the quantity balance, the quantity balance is maintained first. Then, multi-physics field simulation verification tests are performed on multiple initial partitioned battery modules, including thermal simulation, fluid simulation and electrical simulation. Thermal simulation refers to analyzing the uniformity of temperature distribution in each partition, fluid simulation refers to verifying the balance of coolant flow rate, and electrical simulation is to check whether the current distribution is uneven due to partitioning; finally, the partitioning of the initial partitioned battery modules is iteratively optimized, that is, the number of partitions, boundaries or cooling strategies are modified. If the liquid cooling balance does not meet the standards, its priority can be increased and re-partitioned. After three consecutive iterations, if the performance improvement is <2%, it converges and outputs the final M liquid-cooled system partitioned battery modules, that is, the optimized partitioning scheme that meets spatial adjacency, balanced quantity and uniform liquid cooling.

[0036] In step S200 , a multi-sensor network is sequentially arranged on the M liquid cooling system partitioned battery modules, and working data streams of the M partitioned battery modules are collected and acquired through the multi-sensor network.

[0037] Preferably, a multi-sensor network is sequentially deployed on the M liquid cooling system partition battery modules, wherein the multi-sensor network is a distributed monitoring module composed of multiple sensors of different types, which is used to collect multi-dimensional physical quantities reflecting the working status of the battery module in real time. The core sensor deployment data is shown in Table 1:

[0038] Table 1 Multi-sensor network layout data

[0039]

[0040] An edge computing node is formed in each partition to realize data collection and acquisition through partition collection, edge preprocessing, and cloud aggregation. The M liquid cooling system partition battery modules are collected in real time through a multi-sensor sensor network to obtain multi-dimensional monitoring values ​​of parameters such as temperature and voltage for each battery module. The edge node calculates derived indicators in real time, such as temperature gradient, internal resistance change rate, and coolant heat exchange, to form a partition health feature vector. Finally, the working data stream of the M partition battery modules is determined to facilitate the accuracy and reliability of the battery life prediction and health status assessment of the liquid cooling system.

[0041] Step S300: Building a battery module life prediction channel, wherein the battery module life prediction channel includes a conventional battery life prediction branch channel and a personalized battery life prediction branch channel.

[0042] Step S300 further includes step S310, collecting and obtaining a liquid cooling system battery working life data set, wherein the liquid cooling system battery working life data set includes historical liquid cooling battery structure attribute data, battery module working data and corresponding battery life data; step S320, performing cluster analysis on the historical liquid cooling battery structure attribute data according to a preset number of clusters to obtain conventional battery clustering clusters and personalized battery clustering clusters, wherein the preset number of clusters is 2; step S330, classifying and integrating the battery module working data and the corresponding battery life data based on the conventional battery clustering clusters and the personalized battery clustering clusters to obtain a conventional battery life data set and a personalized battery life data set; step S340, performing life prediction training on the conventional battery life data set and the personalized battery life data set to build a battery module life prediction channel.

[0043] Preferably, a liquid-cooled system battery working life data set is collected from a multi-dimensional database formed by the accumulation of long-term monitoring of the operating status of the liquid-cooled battery system, including historical liquid-cooled battery structural attribute data, battery module working data and corresponding battery life data, wherein the historical liquid-cooled battery structural attribute data records the static characteristic parameters of batteries of different batches / models, which may include module geometric dimensions and arrangement, cooling channel design parameters such as flow channel cross-sectional area and pipeline direction, battery cell material properties such as positive and negative electrode formulations and diaphragm types, and thermal conductivity of thermal interface materials; battery module working data refers to the historical dynamic operating parameters of the battery module, including real-time electrical parameters such as voltage, current, internal resistance, temperature field distribution such as maximum / minimum temperature and gradient change, cooling system status such as flow rate and inlet and outlet temperature difference, and mechanical vibration and expansion deformation data; battery life data is the actual aging test result of the corresponding battery module, including capacity attenuation curve, impedance growth trend and marked failure mode.

[0044] Preferably, the K-means algorithm (preset number of clusters K=2) is used to perform cluster analysis on historical liquid-cooled battery structural attribute data, that is, two initial cluster centers, and then weighted Euclidean distance is used to give higher weights to liquid-cooling related attributes, such as cooling contact area, and each liquid-cooled battery structural attribute data sample is divided into the cluster to which the nearest center point belongs according to the distance, and the centroid positions of the two types of clusters are recalculated. When the rate of change of the sum of squared errors within the cluster for two consecutive iterations is <1%, convergence is achieved to obtain conventional battery clustering clusters and personalized battery clustering clusters, among which the conventional battery clustering cluster contains liquid-cooled battery structural attribute data with standard cooling design, accounting for about 70% to 80%; the personalized battery clustering cluster contains liquid-cooled battery structural attribute data with special cooling layout or special-shaped structure, accounting for 20% to 30%.

[0045] Preferably, based on conventional battery clusters and personalized battery clusters, the battery module working data and the corresponding battery life data are classified, integrated and mapped to the corresponding clusters through the battery unique code to obtain a conventional battery life data set and a personalized battery life data set, wherein the conventional battery life data set reflects the typical aging law, including the working parameter sequence of the standard liquid-cooled battery and the corresponding life label; the personalized battery life data set reflects the unique work-life mapping relationship, including the abnormal working mode of the special structure battery and the corresponding accelerated aging data, such as the local accelerated aging caused by asymmetric cooling, the differentiated expansion behavior of the special-shaped battery cells, etc.

[0046] Preferably, life prediction training is performed on conventional battery life datasets and personalized battery life datasets. Specifically, a prediction branch is constructed using a random forest or LSTM neural network, and the conventional battery life dataset is used for training to learn the global aging pattern of liquid cooling system batteries. This obtains a conventional battery life prediction branch channel, which can output a baseline life prediction value based on a standard designed battery module. At the same time, a model is constructed based on a deep neural network, and the personalized battery life dataset is used as input for training. The attention mechanism is combined with the Transformer model to capture local abnormal patterns, and a personalized battery life prediction branch channel is obtained. This can output a life prediction value for the special structure of the liquid cooling system battery module. The conventional battery life prediction branch channel and the personalized battery life prediction branch channel are then integrated to determine the battery module life prediction channel, which is used to make predictions based on the partitioned battery module working data stream and output comprehensive battery module life prediction information.

[0047] Furthermore, step S340 also includes step S341, respectively performing feature identification on the conventional battery life data set and the personalized battery life data set to obtain a conventional battery life sample set and a personalized battery life sample set; step S342, using a deep neural network structure to perform predictive training and iterative optimization on the conventional battery life sample set and the personalized battery life sample set to generate a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; step S343, parallelly fusing the conventional battery life prediction branch channel and the personalized battery life prediction branch channel to build the battery module life prediction channel.

[0048] Preferably, conventional battery life data sets and personalized battery life data sets are extracted respectively, and the most representative features for life prediction are identified. Specifically, for conventional battery life data sets, global aging indicators such as capacity attenuation rate, average internal resistance growth, and number of cycles, steady-state operating parameters such as average charge and discharge temperature, current / voltage fluctuations under standard operating conditions, and cooling performance such as temperature uniformity under nominal flow are extracted, and a conventional battery life sample set is obtained by combination; for personalized battery life data sets, local abnormal features such as the maximum temperature difference between modules, the frequency of tab overheating, and asymmetric expansion are extracted, special operating condition responses such as voltage drop under transient large current and temperature rise rate when cooling fails, as well as structural related parameters such as flow channel design deviation and cell arrangement non-standard coefficient are extracted, and a personalized battery life sample set is determined by combination.

[0049] Preferably, a multi-layer LSTM network is used to construct a prediction branch, which is trained using a conventional battery life sample set. Conventional feature labels such as the number of cycles and average temperature are used as input, and the remaining service life or health status is used as the output label to capture the time dependence in cycle aging. The loss function is then defined by the weighted mean square error to predict battery capacity attenuation, minimizing the prediction deviation under standard working conditions, thereby obtaining a conventional battery life prediction branch channel. A graph neural network and an attention mechanism are used to construct a prediction branch, which is trained using a personalized battery life sample set to process non-uniform spatial distribution features, that is, personalized features such as temperature distribution matrix and flow channel parameters are used as input, and the additional attenuation of the battery module partition is used as the output label to accurately capture the deviation from the baseline aging caused by the special structure, and finally determine the personalized battery life prediction branch channel. Finally, the conventional battery life prediction branch channel and the personalized battery life prediction branch channel are connected in parallel to build a battery module life prediction channel, which can accurately predict the life of the liquid cooling system battery and ensure accuracy.

[0050] Step S400 , mapping the M partitioned battery module working data streams to the battery module life prediction channel for activation prediction analysis to obtain M battery module life prediction information.

[0051] Preferably, the M partitioned battery module working data streams are matched and mapped to the battery module life prediction channel for activation prediction analysis. Specifically, the M partitioned battery module working data streams are time-aligned, and then the liquid cooling system battery characteristics corresponding to the conventional battery life data and personalized battery life data are extracted and the partition type is identified, that is, the historical clustering results are matched based on the partition structure attributes such as flow channel design and cell arrangement. If it belongs to a conventional battery cluster cluster, the conventional battery life prediction branch channel is activated to perform battery life prediction. If it belongs to a personalized battery cluster cluster, the personalized battery life prediction branch channel is activated to perform battery life prediction; finally, differentiated life prediction information for the M battery module partitions is generated, which may include life prediction values ​​based on standard aging, additional life attenuation caused by battery personalized factors, and key influencing factors, such as insufficient cooling efficiency and current concentration.

[0052] Step S500 , performing correlation impact analysis on the life prediction information of the M battery modules according to the distribution characteristic information of the M liquid cooling system partition battery modules, and determining a liquid cooling system battery life prediction result.

[0053] Step S500 further includes step S510, calculating the mean battery module life information of the M battery module life prediction information; step S520, screening and obtaining N battery module life information that does not reach the mean battery module life information in the M battery module life prediction information; step S530, performing correlation impact analysis on the N battery module life information according to the distribution characteristic information of the M liquid cooling system partition battery modules to obtain N battery life impact factors; step S540, performing impact correction on the mean battery module life information based on the N battery life impact factors to determine the liquid cooling system battery life prediction result.

[0054] Preferably, a statistical analysis is performed on the life prediction information such as the remaining number of cycles and health status SOH of all M battery modules, and the arithmetic mean or weighted average of the battery module life is calculated to obtain the mean battery module life information to reflect the expected life benchmark of the liquid-cooled battery system in the absence of local abnormalities; then the M battery module life prediction information is compared with the battery module life mean information, and N battery module life information with a life significantly lower than the battery module life mean information is screened out, wherein N is a positive integer, representing the number of battery modules with lower life prediction values, and N<M, wherein these N battery module partitions usually have problems such as uneven cooling, current overload or mechanical stress concentration, which affect the battery life of the liquid cooling system.

[0055] Preferably, according to the distribution characteristic information of the battery modules in the M liquid cooling system partitions, such as spatial position, cooling flow channel correlation, and electrical connection relationship, a correlation impact analysis is performed on the life information of N battery modules, that is, the impact of the life decline of the substandard battery module partition on the adjacent partitions and the entire liquid cooling system battery is analyzed, which may include but is not limited to thermal diffusion impact, electrical coupling effect and liquid cooling linkage, among which, thermal diffusion impact refers to whether the heat of the high-temperature partition will accelerate the aging of the surrounding modules, the electrical coupling effect refers to whether the increase in the internal resistance of a battery module partition will lead to current redistribution and increase the burden on other battery module partitions, and liquid cooling linkage refers to whether the downstream cooling efficiency is affected when the flow channel of a battery module partition is blocked. By quantifying the correlation effect, a life impact factor is calculated for each substandard battery module partition, that is, N battery life impact factors are obtained, which represent the coefficients that additionally cause the liquid cooling system battery life to be reduced.

[0056] Preferably, N battery life influencing factors are used to modify the average battery module life information. That is, the N battery life influencing factors are integrated into the calculation of the average battery module life information. For example, if the substandard partition is located in the critical cooling path or main current loop, it is given a higher weight. If multiple adjacent partitions fail to meet the standards at the same time, their cumulative impact may cause a nonlinear decline in system life. The final output of the liquid cooling system battery life prediction result reflects both the overall trend and the impact of local anomalies. This allows for precise positioning of high-risk battery module partitions and ensures the accuracy of battery life prediction.

[0057] Furthermore, step S520 also includes step S521, performing correlation impact analysis of the preset spatial radiation distance on each battery module in the N battery module life information in accordance with the distribution characteristic information of the M liquid cooling system partition battery modules, and determining the life attenuation coefficients of the N battery modules; step S522, obtaining the radiation impact quantity of the N battery modules of the N battery module life information; step S523, performing weighted calculation on the life attenuation coefficients of the N battery modules based on the radiation impact quantity of the N battery modules, and obtaining the N battery life impact factors.

[0058] Preferably, for the N battery modules that have been screened out and whose lifespans do not meet the standards, based on the distribution characteristic information of the M liquid cooling system partitioned battery modules, the associated impact analysis of the preset spatial radiation distance is performed on each battery module in turn, wherein the preset spatial radiation distance defines the effective impact range according to the battery type and cooling method, such as the cylindrical battery is set to 3 times the battery cell diameter, and the square battery is set to the distance between two adjacent modules, and then based on the heat radiation conduction, that is, the high-temperature module will increase the surrounding temperature and accelerate the aging of the adjacent area; the current is redistributed, the high internal resistance module causes the current path to change, and increases the load of the specific module; and the cooling liquid flow is disturbed, and the flow channel blockage or leakage will change the downstream cooling efficiency; the radiation impact of each substandard battery module on its surrounding area is comprehensively analyzed, and the associated modules that may be affected within its radiation distance are identified; and the percentage of deviation between the module's own life prediction value and the mean life information of the liquid cooling system battery module is calculated as the corresponding life attenuation coefficient, and the life attenuation coefficients of the N battery modules are determined.

[0059] Preferably, the radiation impact count of N battery modules is obtained, including the total number of associated modules within the direct radiation distance, and the impact weights of common associated modules and key system nodes are distinguished. Then, a weighted calculation is performed based on the number of affected associated modules within the radiation range to amplify the battery module life attenuation coefficient. For example, if one associated module is affected, the life attenuation coefficient is multiplied by 1.2; if three associated modules are affected, the life attenuation coefficient is multiplied by 1.5; if the core cooling branch is affected, the life attenuation coefficient is additionally multiplied by 1.3. Next, a weighted calculation is performed on the N battery module life attenuation coefficients based on the number of radiation impacts of the N battery modules. For example, if battery module A affects three common battery modules, the count is 3; if battery module B affects two common battery modules and one key battery module, the equivalent count is 4. Finally, N battery life impact factors are obtained through two-level weighting, thereby ensuring the accuracy of life prediction for liquid cooling system batteries and improving the reliability of liquid cooling system batteries.

[0060] In the above, refer to Figure 1 The liquid cooling system battery life prediction method according to the embodiment of the present invention is described in detail. Figure 2 A device for predicting battery life of a liquid cooling system according to an embodiment of the present invention is described.

[0061] The liquid cooling system battery life prediction device according to the embodiment of the present invention is used to solve the technical problems existing in the prior art of insufficient battery life prediction accuracy and inability to effectively reflect the non-uniform aging characteristics of liquid cooling system batteries, thereby achieving the technical effects of optimizing battery health management strategies, improving the accuracy of liquid cooling system battery life prediction, and improving the reliability of the liquid cooling system. Figure 2As shown, the liquid cooling system battery life prediction device includes: a battery module partitioning unit 10, a working data stream acquisition unit 20, a life prediction channel building unit 30, an activation prediction analysis unit 40, and an associated impact analysis unit 50.

[0062] The battery module partitioning unit 10 is used to partition the battery modules based on the structural attribute information of the target liquid cooling system battery to obtain M liquid cooling system partitioned battery modules; the working data stream acquisition unit 20 is used to sequentially deploy a multi-sensor sensor network on the M liquid cooling system partitioned battery modules, and collect and obtain the working data streams of the M partitioned battery modules through the multi-sensor sensor network; the life prediction channel construction unit 30 is used to build a battery module life prediction channel, and the battery module life prediction channel includes a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; the activation prediction analysis unit 40 is used to match and map the M partitioned battery module working data streams to the battery module life prediction channel for activation prediction analysis to obtain M battery module life prediction information; the correlation impact analysis unit 50 is used to perform correlation impact analysis on the M battery module life prediction information according to the distribution characteristic information of the M liquid cooling system partitioned battery modules to determine the liquid cooling system battery life prediction result.

[0063] The specific configuration of the battery module partitioning unit 10 will be described in detail below. The battery module partitioning unit 10 further includes: obtaining a battery attribute factor set, wherein the battery attribute factor set includes module layout, arrangement, spatial connection relationship, battery cell size and shape, thermal physical parameters, and liquid cooling integration form; extracting factor parameters from the structural attribute information of the target liquid cooling system battery according to the battery attribute factor set to obtain a liquid cooling battery factor parameter set; constructing a liquid cooling battery partitioning strategy, wherein the liquid cooling battery partitioning strategy includes a battery spatial position adjacency rule, a partition battery quantity balance rule, and a partition battery liquid cooling balance rule; partitioning the battery module based on the liquid cooling battery factor parameter set based on the liquid cooling battery partitioning strategy to obtain M liquid cooling system partition battery modules.

[0064] The specific configuration of the battery module partitioning unit 10 will be described in detail below. The battery module partitioning unit 10 further includes: prioritizing each partitioning rule in the liquid-cooled battery partitioning strategy to determine a liquid-cooled battery partitioning rule sequence; generating a liquid-cooled battery distribution topology map based on the liquid-cooled battery partitioning strategy according to the liquid-cooled battery partitioning rule sequence to obtain multiple initial partitioned battery modules; and performing simulation verification testing and partition iterative optimization on the multiple initial partitioned battery modules to obtain the M liquid-cooled system partitioned battery modules.

[0065] The specific configuration of the life prediction channel building unit 30 will be described in detail below. The life prediction channel building unit 30 further includes: collecting and acquiring a liquid cooling system battery working life data set, wherein the liquid cooling system battery working life data set includes historical liquid cooling battery structure attribute data, battery module working data, and corresponding battery life data; performing cluster analysis on the historical liquid cooling battery structure attribute data according to a preset number of clusters to obtain conventional battery cluster clusters and personalized battery cluster clusters, wherein the preset number of clusters is 2; classifying and integrating the battery module working data and the corresponding battery life data based on the conventional battery cluster clusters and personalized battery cluster clusters to obtain a conventional battery life data set and a personalized battery life data set; performing life prediction training on the conventional battery life data set and the personalized battery life data set to build a battery module life prediction channel.

[0066] The specific configuration of the life prediction channel construction unit 30 will be described in detail below. The life prediction channel construction unit 30 further includes: performing feature identification on the conventional battery life dataset and the personalized battery life dataset to obtain a conventional battery life sample set and a personalized battery life sample set; using a deep neural network structure to perform prediction training and iterative optimization on the conventional battery life sample set and the personalized battery life sample set to generate a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; and parallelly fusing the conventional battery life prediction branch channel and the personalized battery life prediction branch channel to construct the battery module life prediction channel.

[0067] The specific configuration of the correlation impact analysis unit 50 will be described in detail below. The correlation impact analysis unit 50 further includes: calculating the mean battery module life information of the M battery module life prediction information; screening and obtaining N battery module life information from the M battery module life prediction information that does not reach the mean battery module life information; performing correlation impact analysis on the N battery module life information according to the distribution characteristics of the M liquid cooling system partition battery modules to obtain N battery life impact factors; and performing impact correction on the mean battery module life information based on the N battery life impact factors to determine the liquid cooling system battery life prediction result.

[0068] The specific configuration of the correlation impact analysis unit 50 will be described in detail below. The correlation impact analysis unit 50 further includes: performing correlation impact analysis of a preset spatial radiation distance on each of the N battery module life information according to the distribution characteristics of the M liquid cooling system partition battery modules, and determining the life attenuation coefficients of the N battery modules; obtaining the number of N battery module radiation impacts of the N battery module life information; and performing weighted calculation on the N battery module life attenuation coefficients based on the number of N battery module radiation impacts to obtain the N battery life impact factors.

[0069] The liquid cooling system battery life prediction device provided in the embodiment of the present invention can execute the liquid cooling system battery life prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0070] Figure 3 1 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing an embodiment of the present invention. Figure 3 The electronic device shown is merely an example and should not limit the functionality and scope of use of the embodiments of the present invention. The electronic device is implemented as a general-purpose computing device, and its components may include, but are not limited to, an input device 401, a processor 402, a memory 403, and an output device 404. There may be one or more processors 402; the memory 403 may include computer-readable media and at least one program product, which has a set (at least one) of program modules configured to perform the functions of the various embodiments of the present application.

[0071] The memory 403 shown in the embodiment of the present invention may adopt any combination of one or more computer-readable media; the computer-readable storage medium may be, but is not limited to, an infrared, semiconductor system, device or component, or any combination of the above, for storing software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the liquid cooling system battery life prediction method in the embodiment of the present invention. The processor 402 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 403, thereby realizing the above-mentioned liquid cooling system battery life prediction method.

[0072] Based on the above embodiments, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it can implement the liquid cooling system battery life prediction method as described in any of the previous embodiments.

[0073] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0074] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A method for predicting battery life of a liquid cooling system, characterized in that: The method comprises: Partition the battery modules based on the structural attribute information of the target liquid cooling system battery to obtain M liquid cooling system partitioned battery modules; Deploying a multi-sensor network on the M liquid cooling system partitioned battery modules in sequence, and collecting and acquiring working data streams of the M partitioned battery modules through the multi-sensor network; Building a battery module life prediction channel, wherein the battery module life prediction channel includes a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; Matching and mapping the M partitioned battery module working data streams to the battery module life prediction channel to perform activation prediction analysis to obtain M battery module life prediction information; Obtaining distribution characteristic information of the M liquid cooling system partition battery modules, performing correlation impact analysis on the M battery module life prediction information according to the distribution characteristic information of the M liquid cooling system partition battery modules, and determining a liquid cooling system battery life prediction result; The step of establishing a battery module life prediction channel includes: Acquire a liquid cooling system battery service life data set, wherein the liquid cooling system battery service life data set includes historical liquid cooling battery structure attribute data, battery module operation data, and corresponding battery life data; Performing cluster analysis on the historical liquid-cooled battery structural attribute data according to a preset number of clusters to obtain conventional battery clusters and personalized battery clusters, wherein the preset number of clusters is 2; Classifying and integrating the battery module operating data and corresponding battery life data based on the conventional battery clusters and the personalized battery clusters to obtain a conventional battery life data set and a personalized battery life data set; Performing life prediction training on the conventional battery life dataset and the personalized battery life dataset to build a battery module life prediction channel; The step of determining the predicted result of the battery life of the liquid cooling system includes: Calculating battery module life average information of the M battery module life prediction information; Filtering and obtaining N battery module life information that does not reach the battery module life mean information among the M battery module life prediction information; Performing a correlation impact analysis on the life information of the N battery modules according to the distribution characteristic information of the M liquid cooling system partition battery modules to obtain N battery life impact factors; Based on the N battery life influencing factors, the battery module life mean information is corrected to determine the liquid cooling system battery life prediction result.

2. The liquid cooling system battery life prediction method according to claim 1, characterized in that: The M liquid cooling system partitioned battery modules are obtained, including: Obtaining a set of battery attribute factors, the set of battery attribute factors including module layout, arrangement, spatial connection relationship, battery cell size and shape, thermal physical parameters, and liquid cooling integration form; Extracting factor parameters from the structural attribute information of the target liquid cooling system battery according to the battery attribute factor set to obtain a liquid cooling battery factor parameter set; Constructing a liquid-cooled battery partitioning strategy, wherein the liquid-cooled battery partitioning strategy includes a battery spatial location adjacency rule, a partition battery quantity balance rule, and a partition battery liquid cooling balance rule; The liquid-cooled battery factor parameter set is partitioned into battery modules based on the liquid-cooled battery partitioning strategy to obtain M liquid-cooled system partitioned battery modules.

3. The liquid cooling system battery life prediction method according to claim 2, characterized in that: The method of obtaining M liquid cooling system partitioned battery modules includes: Prioritizing each partitioning rule in the liquid-cooled battery partitioning strategy to determine a liquid-cooled battery partitioning rule sequence; generating a liquid-cooled battery distribution topology map according to the liquid-cooled battery factor parameter set; Partitioning the liquid-cooled battery distribution topology map into battery modules based on the liquid-cooled battery partitioning strategy according to the liquid-cooled battery partitioning rule sequence to obtain a plurality of initial partitioned battery modules; The multiple initial partitioned battery modules are subjected to simulation verification testing and partition iterative optimization to obtain the M liquid cooling system partitioned battery modules.

4. The liquid cooling system battery life prediction method according to claim 1, characterized in that: The building of a battery module life prediction channel includes: Performing feature identification on the conventional battery life data set and the personalized battery life data set respectively to obtain a conventional battery life sample set and a personalized battery life sample set; Using a deep neural network structure to perform prediction training and iterative optimization on the conventional battery life sample set and the personalized battery life sample set to generate a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; The conventional battery life prediction branch channel and the personalized battery life prediction branch channel are connected in parallel and integrated to build the battery module life prediction channel.

5. The liquid cooling system battery life prediction method according to claim 1, characterized in that: The obtained N battery life influencing factors include: Performing a correlation impact analysis of a preset spatial radiation distance on each battery module in the N battery module life information in accordance with the distribution characteristic information of the M liquid cooling system partition battery modules, and determining the life attenuation coefficient of the N battery modules; Obtaining the radiation impact quantity of the N battery modules of the life information of the N battery modules; The N battery life impact factors are obtained by performing weighted calculation on the life attenuation coefficients of the N battery modules based on the number of radiation impacts of the N battery modules.

6. A liquid cooling system battery life prediction device, characterized in that: The device is used to implement the liquid cooling system battery life prediction method according to any one of claims 1 to 5, and the device includes: A battery module partitioning unit is used to partition the battery modules based on the structural attribute information of the target liquid cooling system battery to obtain M liquid cooling system partitioned battery modules; A working data stream acquisition unit is used to sequentially deploy a multi-sensor sensor network on the M liquid cooling system partition battery modules, and collect and acquire the working data streams of the M partition battery modules through the multi-sensor sensor network; A life prediction channel building unit, used to build a battery module life prediction channel, wherein the battery module life prediction channel includes a conventional battery life prediction branch channel and a personalized battery life prediction branch channel; An activation prediction analysis unit is used to match and map the M partitioned battery module working data streams to the battery module life prediction channel to perform activation prediction analysis to obtain M battery module life prediction information; The correlation impact analysis unit is used to perform correlation impact analysis on the life prediction information of the M battery modules according to the distribution characteristic information of the M liquid cooling system partition battery modules, and determine the liquid cooling system battery life prediction result.

7. An electronic device, characterized in that: The electronic device comprises: a memory for storing executable instructions; The processor is configured to implement the liquid cooling system battery life prediction method according to any one of claims 1 to 5 when executing the executable instructions stored in the memory.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the liquid cooling system battery life prediction method according to any one of claims 1 to 5 is implemented.

Images