Battery reliability evaluation method, device and equipment and storage medium

By using a CNN-LSTM model to jointly estimate the state of charge, state of health, and temperature state of lithium-ion batteries, the problem of complex and inefficient assessment in existing technologies is solved, achieving efficient and accurate assessment of battery state and ensuring the safe and stable operation of batteries.

CN118938018APending Publication Date: 2024-11-12STATE GRID ZHEJIANG ELECTRIC POWER CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411069416.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In existing technologies, reliability assessment methods for lithium-ion batteries typically only consider a single state factor, ignoring the state of charge, state of health, and temperature state. This results in complex and inefficient assessments, and makes it impossible to achieve coordinated estimation of multiple states.

Method used

A CNN-LSTM-based model is used to perform joint estimation of state of charge, state of health, and temperature state by acquiring battery measurement data and fuzzy health values. Data features are extracted using convolutional and LSTM layers to generate state values, which are then quantified using piecewise functions and state thresholds to obtain a comprehensive reliability assessment value for the battery.

Benefits of technology

It achieves efficient collaborative assessment of battery status, improves assessment efficiency and accuracy, and can comprehensively assess changes in battery reliability, guiding the safe and stable operation and maintenance of batteries.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938018A_ABST
    Figure CN118938018A_ABST
Patent Text Reader

Abstract

The invention discloses a battery reliability assessment method and device, equipment and a storage medium. The method comprises the steps of obtaining measurement data and a fuzzy health value of a to-be-assessed battery; wherein the measurement data comprises voltage, current, constant-current charging time and constant-current discharging time; inputting the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain a charge state value, a health state value and a temperature state value output by the model; and respectively quantifying the charge state value, the health state value and the temperature state value to obtain corresponding reliability quantized values, and determining a reliability evaluation value of the to-be-evaluated battery according to the reliability quantized values. According to the method, a complex physical model does not need to be established, collaborative estimation of the charge state, the health state and the temperature state of the battery is realized, and the calculation efficiency can be improved while the reliability evaluation accuracy of the battery is ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of battery status detection, and in particular to a battery reliability evaluation method, device, equipment and storage medium. Background Art

[0002] Lithium-ion batteries have the advantages of high energy density, long cycle life, and low self-discharge rate, and are widely used around the world. However, during the use of lithium-ion batteries, very complex chemical reactions will occur inside them, which may cause thermal runaway at high temperatures or in overcharged states, leading to battery fire or explosion. Therefore, in order to avoid accidents such as overcharging and discharging, excessive aging, and thermal runaway of batteries, and to ensure the safe and stable operation of batteries, it is very important to monitor various battery states in real time and perform reliability analysis on batteries. For the reliability assessment of battery equipment, it is necessary to comprehensively consider the various aspects of the battery state in order to give reasonable and effective reliability assessment results, thereby guiding the operation and maintenance of battery equipment. However, existing battery reliability assessment methods often only consider a single factor of battery SOH (State of health), and states such as SOC (State of charge) and SOT (State of temperature) are ignored in reliability assessment. In the prior art, if multiple states of a battery are to be evaluated, it is usually necessary to establish multiple physical models to evaluate each state separately. The modeling is complex and the calculation is large, the battery detection efficiency is low, and it is impossible to achieve coordinated estimation and reliability assessment of multiple battery states. Summary of the invention

[0003] To solve the above technical problems, the present invention provides a battery reliability assessment method, device, equipment and storage medium, which can realize the coordinated estimation and reliability assessment of multiple battery states without the need to establish a complex physical model, and can improve the computing efficiency while ensuring the accuracy of battery reliability assessment.

[0004] An embodiment of the present invention provides a battery reliability assessment method, comprising:

[0005] Obtaining measurement data and a fuzzy health value of the battery to be evaluated; wherein the measurement data includes voltage, current, constant current charging time, and constant current discharging time;

[0006] Inputting the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain the state of charge value, health state value and temperature state value of the battery to be evaluated output by the model; wherein the CNN-LSTM model is trained based on historical measurement data and historical fuzzy health values;

[0007] The state of charge value, the health state value, and the temperature state value are quantified respectively to obtain corresponding reliability quantization values, and the reliability evaluation value of the battery to be evaluated is determined according to the reliability quantization values.

[0008] As an improvement of the above solution, the fuzzy health value is calculated by the following formula:

[0009]

[0010] Among them, SOH * is the fuzzy health value, SOH(0) is the initial health value, N is the number of cycles, DOD is the depth of discharge, and a and b are preset parameter values.

[0011] As an improvement of the above solution, the CNN-LSTM model is a mapping model of battery data and battery status, and the CNN-LSTM model includes:

[0012] An input layer, used for receiving the measurement data and the fuzzy health value;

[0013] Convolutional layer, used to extract data features and input the features into the LSTM layer;

[0014] The LSTM layer is used to analyze the temporal characteristics of the data and generate state of charge, state of health, and temperature values.

[0015] As an improvement to the above solution, the state of charge value is quantified, specifically including:

[0016] When the state of charge value is less than 1, setting the reliability quantization value of the state of charge value to 1;

[0017] When the state of charge value is greater than 1, a ratio of a difference between the state of charge value and 1 and a preset overcharge threshold is used as a reliability quantification value of the state of charge value.

[0018] As an improvement to the above solution, the health status value is quantified, specifically including:

[0019] When the health status value is greater than a preset health status threshold, the health status value is used as a reliability quantification value of the health status value;

[0020] When the health status value is less than the health status threshold, the reliability quantization value of the health status value is set to 0.

[0021] As an improvement to the above solution, the temperature state value is quantified, specifically including:

[0022] When the temperature state value is less than a preset temperature threshold, taking the ratio of the difference between the temperature threshold and the temperature state value to the temperature threshold as a reliability quantification value of the temperature state value;

[0023] When the temperature state value is greater than the temperature threshold, the reliability quantization value of the temperature state value is set to 0.

[0024] As an improvement of the above solution, the formula for the reliability evaluation value of the battery to be evaluated is:

[0025] R cell =min(R c , R h , R t );

[0026] Among them, R cell is the comprehensive reliability evaluation value of the battery, R c is the reliability quantification value of the battery state of charge, R h is the reliability quantification value of the battery health status, R t A reliability quantification value for the battery temperature state.

[0027] The embodiment of the present invention further provides a battery reliability assessment device, comprising:

[0028] A data acquisition module, used to acquire measurement data and fuzzy health value of the battery to be evaluated; wherein the measurement data includes voltage, current, constant current charging time and constant current discharging time;

[0029] A state collaborative estimation module, used for inputting the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain the state of charge value, health state value and temperature state value of the battery to be evaluated output by the model; wherein the CNN-LSTM model is trained based on historical measurement data and historical fuzzy health values;

[0030] The reliability evaluation module is used to quantify the state of charge value, the health state value and the temperature state value respectively to obtain corresponding reliability quantification values, and determine the reliability evaluation value of the battery to be evaluated according to the reliability quantification values.

[0031] An embodiment of the present invention further provides a terminal device, including a processor and a memory, wherein a computer program is stored in the memory, and the computer program is configured to be executed by the processor, and when the processor executes the computer program, the battery reliability assessment method described in any one of the above items is implemented.

[0032] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned battery reliability assessment methods.

[0033] Compared with the prior art, the battery reliability assessment method, device, equipment and storage medium provided by the embodiment of the present invention have the following beneficial effects: by establishing a battery state of charge, health state and temperature state collaborative estimation model based on CNN-LSTM, the problem of needing to establish multiple models when evaluating multiple states of the battery is solved, and there is no need to establish a physical model, which improves the efficiency of battery state monitoring and can more comprehensively evaluate the battery state; by calculating the fuzzy health value and using it as the input of the CNN-LSTM model, the accuracy of the model is guaranteed; starting from the mechanism of battery reliability degradation, a piecewise function and a state threshold are established, and the state of charge value, health state value and temperature state value obtained by collaborative estimation are quantified respectively, so as to obtain the comprehensive reliability of the battery, solve the problem that battery reliability is difficult to quantify and evaluate, improve the effectiveness of battery reliability evaluation, and can effectively and comprehensively evaluate the reliability changes during battery operation. The embodiment of the present invention realizes the collaborative estimation of battery state of charge, health state and temperature state, and can effectively and comprehensively evaluate battery reliability to guide the operation and maintenance of battery equipment and ensure the safe and stable operation of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flow chart of a battery reliability assessment method provided by an embodiment of the present invention;

[0035] Figure 2 This is a collaborative evaluation result diagram of a CNN-LSTM model in a battery reliability evaluation method provided by an embodiment of the present invention;

[0036] Figure 3 is a battery reliability evaluation result diagram of a battery reliability evaluation method provided by an embodiment of the present invention;

[0037] Figure 4 is a structural schematic diagram of a battery reliability assessment device provided by an embodiment of the present invention;

[0038] Figure 5 It is a structural diagram of a terminal device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments 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 creative work are within the scope of protection of the present invention.

[0040] See also Figure 1 , Figure 1 1 is a flow chart of a battery reliability evaluation method provided by an embodiment of the present invention. The battery reliability evaluation method comprises:

[0041] S1: Obtaining measurement data and fuzzy health value of the battery to be evaluated; wherein the measurement data includes voltage, current, constant current charging time and constant current discharging time;

[0042] S2: Inputting the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain the state of charge value, health state value and temperature state value of the battery to be evaluated output by the model; wherein the CNN-LSTM model is trained based on historical measurement data and historical fuzzy health values;

[0043] S3: quantifying the state of charge value, the health state value, and the temperature state value respectively to obtain corresponding reliability quantization values, and determining the reliability evaluation value of the battery to be evaluated according to the reliability quantization values.

[0044] Specifically, the embodiment of the present invention adopts a data-driven method to evaluate the battery state. The data-driven method extracts data features that are strongly correlated with the battery state from external measurable parameters, and then establishes a mapping model between the data features and the battery state through algorithms such as machine learning. However, there are few existing data-driven methods that can collaboratively evaluate multiple battery states. In this regard, the embodiment of the present invention provides a battery state of charge, health state and temperature state collaborative estimation model based on CNN-LSTM (Convolutional neural network with long short-term memory network, convolutional neural network and long short-term memory neural network), and the obtained fuzzy health value and the measured battery voltage, current, constant current charging time and constant current discharge time are input into the trained CNN-LSTM model to achieve collaborative estimation of the battery state of charge, health state and temperature state, and then determine the reliability evaluation value of the battery to be evaluated. The present invention does not need to establish multiple state evaluation models, nor does it need to establish a complex physical model, and can improve the evaluation efficiency while ensuring accuracy.

[0045] As an optional embodiment, the fuzzy health value is calculated by the following formula:

[0046]

[0047] Among them, SOH * is the fuzzy health value, SOH(0) is the initial health value, N is the number of cycles, DOD is the depth of discharge, and a and b are preset parameter values.

[0048] Specifically, a and b are parameter values ​​obtained by fitting the battery operation data, and a and b are preferably set to 694 and 0.795, and then the fuzzy health value of the battery is calculated according to the initial health value, cycle number and discharge depth of the battery. The present invention ensures the accuracy of model evaluation by using the fuzzy health value as the input of the CNN-LSTM model.

[0049] Furthermore, the voltage, current, constant current charging time, constant current discharging time and fuzzy health value of the battery to be evaluated are concatenated to obtain the input vector of the model, and then input into the trained CNN-LSTM model for coordinated evaluation of the three state quantities of the battery to obtain the battery state of charge value, health state value and temperature state value output by the model.

[0050] As one of the optional embodiments, the CNN-LSTM model is a mapping model of battery data and battery status, and the CNN-LSTM model includes:

[0051] An input layer, used for receiving the measurement data and the fuzzy health value;

[0052] Convolutional layer, used to extract data features and input the features into the LSTM layer;

[0053] The LSTM layer is used to analyze the temporal characteristics of the data and generate state of charge, state of health, and temperature values.

[0054] Specifically, the CNN-LSTM model is:

[0055]

[0056] i t =σ(W xi x t +W hi h t-1 +b i );

[0057] f t =σ(W xf x t +W hf h t-1 +bf );

[0058] o t =v(W xo x t +W ho h t-1 +b o );

[0059]

[0060] h t =o t ⊙tanh(c t );

[0061] Among them, X is the input data of the model, x t is the output of CNN at time t, f is the activation function of CNN, represents the convolution operation, W is the weight vector of CNN, b is the bias term of CNN, i t 、f t and t are the input gate, forget gate and output gate of LSTM at time t, σ is the sigmoid activation function, W xi , W xf , W xo and W xc are the weight vectors of the input gate, forget gate, output gate and memory unit of LSTM respectively, W hi , W hf , W ho and W hc are the weight vectors between the hidden layer of LSTM and the input gate, forget gate, output gate, and memory unit, respectively. i , b f , b o and b c They are the input gate, forget gate, output gate and bias term of memory unit of LSTM respectively. is the candidate memory unit of LSTM at time t, c t is the memory unit of LSTM at time t, c t-1 is the memory unit of LSTM at time t-1, ⊙ represents the vector product operation, and h t is the output result of the CNN-LSTM model at time t, h t-1 is the output result of the CNN-LSTM model at time t-1, and tanh is the hyperbolic tangent activation function.

[0062] The model output results include the battery's state of charge, health status, and temperature status.

[0063] Furthermore, in response to the problem that battery reliability is difficult to quantify and evaluate, the embodiments of the present invention start from the mechanism of battery reliability degradation, and according to the charge state value, health state value and temperature state value evaluated by the CNN-LSTM model, set corresponding piecewise functions and state thresholds to quantify the battery reliability affected by the charge state, health state and temperature state, and thus obtain the comprehensive battery reliability.

[0064] As one of the optional embodiments, quantifying the state of charge value specifically includes:

[0065] When the state of charge value is less than 1, setting the reliability quantization value of the state of charge value to 1;

[0066] When the state of charge value is greater than 1, a ratio of a difference between the state of charge value and 1 and a preset overcharge threshold is used as a reliability quantification value of the state of charge value.

[0067] Specifically, for the state of charge of the battery, when the state of charge value is less than 1, the battery is considered to be normally charged, and its reliability quantification value is 1; when the state of charge value is greater than 1, the battery will be overcharged, thereby reducing reliability. Therefore, the state of charge value is quantified according to the state of charge value and the overcharge threshold to obtain the corresponding reliability quantification value. The reliability quantification value of the battery state of charge represents the battery reliability considering only the battery state of charge. Among them, the quantification formula of the state of charge value is:

[0068]

[0069] Among them, R c is the reliability quantification value of the battery state of charge, SOC is the state of charge value, v c is the overcharge threshold of the battery. Preferably, the overcharge threshold v c The preferred setting is 0.2.

[0070] As one of the optional embodiments, quantifying the health status value specifically includes:

[0071] When the health status value is greater than a preset health status threshold, the health status value is used as a reliability quantification value of the health status value;

[0072] When the health status value is less than the health status threshold, the reliability quantization value of the health status value is set to 0.

[0073] Specifically, for the health status of the battery, when the health status value is less than the health status threshold, the battery is generally considered to be aged and scrapped, and its reliability quantization value is set to 0; when the health status value is greater than the health status threshold, the corresponding reliability quantization value is obtained according to the health status threshold, and the reliability quantization value of the battery health status represents the battery reliability considering only the battery health status. Among them, the quantification formula of the health status value is:

[0074]

[0075] Among them, R h is the reliability quantification value of the battery health status, SOH is the health status value, v h is the health status threshold. Preferably, the health status threshold v h The preferred setting is 0.8.

[0076] As one of the optional embodiments, quantifying the temperature state value specifically includes:

[0077] When the temperature state value is less than a preset temperature threshold, taking the ratio of the difference between the temperature threshold and the temperature state value to the temperature threshold as a reliability quantification value of the temperature state value;

[0078] When the temperature state value is greater than the temperature threshold, the reliability quantization value of the temperature state value is set to 0.

[0079] Specifically, for the temperature state of the battery, when the temperature state value is greater than the temperature threshold, the battery is generally considered scrapped, and its reliability quantization value is set to 0; when the temperature state value is less than the temperature threshold, the temperature state value is quantified according to the temperature threshold and the temperature state value to obtain the corresponding reliability quantization value. The reliability quantization value of the battery temperature state represents the battery reliability considering only the battery temperature state. Among them, the quantization formula of the temperature state value is:

[0080]

[0081] Among them, R t is the reliability quantification value of the battery temperature state, SOT is the temperature state value, v t Preferably, the temperature state threshold v t The preferred setting is 80.

[0082] Furthermore, the formula for the reliability evaluation value of the battery to be evaluated is:

[0083] R cell =min(R c , R h , R t );

[0084] Among them, R cell is the comprehensive reliability evaluation value of the battery, R c is the reliability quantification value of the battery state of charge, R h is the reliability quantification value of the battery health status, R t A reliability quantification value for the battery temperature state.

[0085] Specifically, the comprehensive reliability evaluation value of the battery is determined according to the minimum value of the reliability quantization values ​​corresponding to the three state values. Based on the battery reliability evaluation method provided by the embodiment of the present invention, the battery state of charge, temperature state, health state and comprehensive reliability state can be obtained. The evaluation values ​​of the corresponding states can be selected according to actual needs to evaluate the battery performance from different angles, providing guidance on whether the battery can operate reliably.

[0086] In order to illustrate the technical effect of the battery reliability assessment method provided by the embodiment of the present invention, the battery data numbered B0007 in the NASA public data set is selected for CNN-LSTM model training, wherein some of the battery data are shown in Table 1.

[0087] Table 1 Part of NASA battery data

[0088] Time / s Current / A Voltage / V Constant current charging time / s Constant current discharge time / s Fuzzy health value 1 0.094 4.267 10805 2837 0.98 500 0.0577 4.250 10805 2837 0.98 1000 0.0332 4.240 10805 2837 0.98 1500 -0.0012 0.003 10805 2837 0.98 2000 1.5116 4.742 10038 2839 0.96

[0089] Use the trained CNN-LSTM model to evaluate the reliability of the battery. Figure 2 , Figure 2 This is a collaborative evaluation result diagram of the CNN-LSTM model in a battery reliability evaluation method provided by an embodiment of the present invention, which shows the estimated errors of the battery state of charge value SOC, health state value SOH and temperature state value SOT obtained by the model, and their average absolute errors are 1.47%, 4.8% and 0.55℃ respectively. Please refer to Figure 3 , Figure 3 This is a battery comprehensive reliability evaluation result diagram of a battery reliability evaluation method provided by an embodiment of the present invention, which shows the change of battery reliability over time. It can be seen from the figure that the battery reliability evaluation method proposed by the present invention can effectively evaluate battery reliability and reflect the reliability changes caused by changes in battery charge state, health state and temperature state, which improves efficiency while ensuring accuracy.

[0090] The embodiment of the present invention solves the problem of needing to establish multiple models when evaluating multiple states of the battery by establishing a collaborative estimation model of the battery state of charge, health state and temperature state based on CNN-LSTM, and does not need to establish a physical model, thereby improving the efficiency of battery state monitoring and being able to more comprehensively evaluate the battery state; by calculating the fuzzy health value and using it as the input of the CNN-LSTM model, the accuracy of the model is guaranteed; starting from the mechanism of battery reliability degradation, a piecewise function and a state threshold are established, and the state of charge value, health state value and temperature state value obtained by collaborative estimation are quantified respectively, thereby obtaining the comprehensive reliability of the battery, solving the problem of difficulty in quantitative evaluation of battery reliability, improving the effectiveness of battery reliability evaluation, and being able to effectively and comprehensively evaluate the reliability changes during battery operation. The embodiment of the present invention realizes the collaborative estimation of the battery state of charge, health state and temperature state, and can effectively and comprehensively evaluate the battery reliability to guide the operation and maintenance of battery equipment and ensure the safe and stable operation of the battery.

[0091] Correspondingly, the present invention also provides a battery reliability evaluation device, which can implement all the processes of the battery reliability evaluation method in the above embodiment.

[0092] See also Figure 4 , Figure 4 1 is a schematic diagram of a battery reliability assessment device provided by an embodiment of the present invention. The battery reliability assessment device comprises:

[0093] The data acquisition module 401 is used to acquire the measurement data and fuzzy health value of the battery to be evaluated; wherein the measurement data includes voltage, current, constant current charging time and constant current discharging time;

[0094] The state collaborative estimation module 402 is used to input the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain the state of charge value, health state value and temperature state value of the battery to be evaluated output by the model; wherein the CNN-LSTM model is trained based on historical measurement data and historical fuzzy health values;

[0095] The reliability evaluation module 403 is used to quantify the state of charge value, the health state value and the temperature state value respectively to obtain corresponding reliability quantization values, and determine the reliability evaluation value of the battery to be evaluated according to the reliability quantization values.

[0096] Preferably, the fuzzy health value is calculated by the following formula:

[0097]

[0098] Among them, SOH *is the fuzzy health value, SOH(0) is the initial health value, N is the number of cycles, DOD is the depth of discharge, and a and b are preset parameter values.

[0099] Preferably, the CNN-LSTM model is a mapping model of battery data and battery status, and the CNN-LSTM model includes:

[0100] An input layer, used for receiving the measurement data and the fuzzy health value;

[0101] Convolutional layer, used to extract data features and input the features into the LSTM layer;

[0102] The LSTM layer is used to analyze the temporal characteristics of the data and generate state of charge, state of health, and temperature values.

[0103] Preferably, quantifying the state of charge value specifically includes:

[0104] When the state of charge value is less than 1, setting the reliability quantization value of the state of charge value to 1;

[0105] When the state of charge value is greater than 1, a ratio of a difference between the state of charge value and 1 and a preset overcharge threshold is used as a reliability quantification value of the state of charge value.

[0106] Preferably, quantifying the health status value specifically includes:

[0107] When the health status value is greater than a preset health status threshold, the health status value is used as a reliability quantification value of the health status value;

[0108] When the health status value is less than the health status threshold, the reliability quantization value of the health status value is set to 0.

[0109] Preferably, quantifying the temperature state value specifically includes:

[0110] When the temperature state value is less than a preset temperature threshold, taking the ratio of the difference between the temperature threshold and the temperature state value to the temperature threshold as a reliability quantification value of the temperature state value;

[0111] When the temperature state value is greater than the temperature threshold, the reliability quantization value of the temperature state value is set to 0.

[0112] Preferably, the formula for the reliability evaluation value of the battery to be evaluated is:

[0113] R cell =min(R c , R h , R t);

[0114] Among them, R cell is the comprehensive reliability evaluation value of the battery, R c is the reliability quantification value of the battery state of charge, R h is the reliability quantification value of the battery health status, R t A reliability quantification value for the battery temperature state.

[0115] In specific implementation, the working principle, control process and technical effect achieved by the battery reliability assessment device provided in the embodiment of the present invention are the same as those of the battery reliability assessment method in the above embodiment, and will not be repeated here.

[0116] See also Figure 5 , Figure 5 1 is a schematic diagram of a terminal device provided in an embodiment of the present invention. The terminal device includes a processor 501, a memory 502, and a computer program stored in the memory 502 and configured to be executed by the processor 501. When the processor 501 executes the computer program, the battery reliability evaluation method described in any of the above embodiments is implemented.

[0117] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory 502 and executed by the processor 501 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0118] The processor 501 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor, or the processor 501 can be any conventional processor. The processor 501 is the control center of the terminal device, and various parts of the terminal device are connected using various interfaces and lines.

[0119] The memory 502 mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 502 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory 502 can also be other volatile solid-state storage devices.

[0120] It should be noted that the above terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that Figure 5 The structural diagram is only an example of the above-mentioned terminal device and does not constitute a limitation of the above-mentioned terminal device. It may include more or less components than shown in the figure, or combine certain components, or different components.

[0121] An embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the battery reliability assessment method described in any of the above embodiments.

[0122] The beneficial effects of a battery reliability assessment method, device, equipment and storage medium provided by the embodiment of the present invention are: by establishing a battery state of charge, health state and temperature state collaborative estimation model based on CNN-LSTM, the problem of needing to establish multiple models when evaluating multiple states of the battery is solved, and there is no need to establish a physical model, which improves the efficiency of battery state monitoring and can more comprehensively evaluate the battery state; by calculating the fuzzy health value and using it as the input of the CNN-LSTM model, the accuracy of the model is guaranteed; starting from the mechanism of battery reliability degradation, a piecewise function and a state threshold are established, and the state of charge value, health state value and temperature state value obtained by collaborative estimation are quantified respectively, so as to obtain the comprehensive reliability of the battery, solve the problem that battery reliability is difficult to quantify and evaluate, improve the effectiveness of battery reliability evaluation, and can effectively and comprehensively evaluate the reliability changes during battery operation. The embodiment of the present invention realizes the collaborative estimation of battery state of charge, health state and temperature state, and can effectively and comprehensively evaluate battery reliability to guide the operation and maintenance of battery equipment and ensure the safe and stable operation of the battery.

[0123] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A battery reliability assessment method, characterized in that: include: Obtaining measurement data and a fuzzy health value of the battery to be evaluated; wherein the measurement data includes voltage, current, constant current charging time, and constant current discharging time; Inputting the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain the state of charge value, health state value and temperature state value of the battery to be evaluated output by the model; wherein the CNN-LSTM model is trained based on historical measurement data and historical fuzzy health values; The state of charge value, the health state value, and the temperature state value are quantified respectively to obtain corresponding reliability quantization values, and the reliability evaluation value of the battery to be evaluated is determined according to the reliability quantization values.

2. The battery reliability evaluation method according to claim 1, characterized in that: The fuzzy health value is calculated by the following formula: Among them, SOH * is the fuzzy health value, SOH(0) is the initial health value, N is the number of cycles, DOD is the depth of discharge, and a and b are preset parameter values.

3. The battery reliability evaluation method according to claim 1, characterized in that: The CNN-LSTM model is a mapping model between battery data and battery status, and the CNN-LSTM model includes: An input layer, used for receiving the measurement data and the fuzzy health value; Convolutional layer, used to extract data features and input the features into the LSTM layer; The LSTM layer is used to analyze the temporal characteristics of the data and generate state of charge, state of health, and temperature values.

4. The battery reliability evaluation method according to claim 1, wherein: Quantifying the state of charge value specifically includes: When the state of charge value is less than 1, setting the reliability quantization value of the state of charge value to 1; When the state of charge value is greater than 1, a ratio of a difference between the state of charge value and 1 and a preset overcharge threshold is used as a reliability quantification value of the state of charge value.

5. The battery reliability evaluation method according to claim 4, characterized in that: Quantifying the health status value specifically includes: When the health status value is greater than a preset health status threshold, the health status value is used as a reliability quantification value of the health status value; When the health status value is less than the health status threshold, the reliability quantization value of the health status value is set to 0.

6. The battery reliability evaluation method according to claim 5, characterized in that: Quantifying the temperature state value specifically includes: When the temperature state value is less than a preset temperature threshold, taking the ratio of the difference between the temperature threshold and the temperature state value to the temperature threshold as a reliability quantification value of the temperature state value; When the temperature state value is greater than the temperature threshold, the reliability quantization value of the temperature state value is set to 0.

7. The battery reliability evaluation method according to claim 6, characterized in that: The formula for the reliability evaluation value of the battery to be evaluated is: R cell =min(R c ,R h ,R t ); Among them, R cell is the comprehensive reliability evaluation value of the battery, R c is the reliability quantification value of the battery state of charge, R h is the reliability quantification value of the battery health status, R t A reliability quantification value for the battery temperature state.

8. A battery reliability assessment device, characterized in that: include: A data acquisition module, used to acquire measurement data and fuzzy health value of the battery to be evaluated; wherein the measurement data includes voltage, current, constant current charging time and constant current discharging time; A state collaborative estimation module, used for inputting the measurement data and the fuzzy health value into a preset CNN-LSTM model to obtain the state of charge value, health state value and temperature state value of the battery to be evaluated output by the model; wherein the CNN-LSTM model is trained based on historical measurement data and historical fuzzy health values; The reliability evaluation module is used to quantify the state of charge value, the health state value and the temperature state value respectively to obtain corresponding reliability quantification values, and determine the reliability evaluation value of the battery to be evaluated according to the reliability quantification values.

9. A terminal device, characterized in that: The invention comprises a processor and a memory, wherein a computer program is stored in the memory and the computer program is configured to be executed by the processor, and when the processor executes the computer program, the battery reliability evaluation method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the battery reliability evaluation method according to any one of claims 1 to 7 is implemented.

Citation Information

Cited By

  • Battery energy storage power station state evaluation and intelligent sensing method and system

    CN120065000A

  • Battery health assessment method and device based on three-dimensional correction model, and terminal

    CN121276377A

  • Battery health evaluation method and device based on three-dimensional correction model and terminal

    CN121276377B