Life prediction method, system and device based on ultra-capacitor coupled lithium battery energy storage

Abstract

The present invention discloses a life prediction method, system and device based on ultracapacitor coupled lithium battery energy storage. First, multiple groups of input variables are selected according to the actual application scenario, and these variables are preprocessed to obtain preprocessed input variables; secondly, based on the preprocessed input variables, the linearization degree between any two groups of variables is evaluated, and the variables with high linearization degree are linearly fitted to obtain multiple groups of fitting variables; then, the influence effect of these fitting variables is corrected to obtain corrected variables, which will more accurately reflect the factors affecting the battery life; finally, the multiple groups of corrected variables are input into the prediction model for processing to obtain the predicted data of the battery life. Not only the on-site conditions of battery use are taken into account, but also the accuracy of the prediction is improved through linear fitting and correction of influence effects, providing a more accurate and effective technical means for the prediction of battery life.

Description

Technical Field

[0001] The present invention belongs to the technical field of hybrid energy storage, and relates to a life prediction method, system and device based on ultra-capacitor coupled lithium battery energy storage. Background Art

[0002] Thermal energy storage combined frequency regulation can significantly improve the frequency regulation performance of thermal power units and can quickly and effectively reduce the system frequency regulation capacity shortage. At present, thermal energy storage combined frequency regulation technologies include battery energy storage, supercapacitor energy storage, flywheel energy storage, molten salt energy storage and mixed energy storage in various forms; battery energy storage has a low cycle life and certain safety hazards. Supercapacitor energy storage and flywheel energy storage, as representatives of power-type energy storage devices, have defects such as high cost and low energy density. Molten salt energy storage uses raw materials such as nitrate as heat storage medium, and stores and releases energy through the conversion of heat energy of heat transfer medium and internal energy of molten salt. It has the advantages of low cost, high safety, large capacity and long life, but molten salt energy storage has the risk of salt corrosion and crystallization.

[0003] Supercapacitor-coupled lithium battery energy storage has broad application prospects in many fields. This hybrid energy storage method can extend the service life of the battery. That is, the fast charging and discharging capability of the supercapacitor can balance the charging and discharging process of the lithium battery, reduce the charging and discharging pressure of the lithium battery, and thus extend the service life of the lithium battery.

[0004] The number of times a lithium battery is used is its service life. The inputs of existing lithium battery life prediction models are mostly internal resistance, voltage, current, and temperature. However, there are many factors that affect the life of lithium batteries. Some factors will affect each other, some factors have a large impact, and some factors have a small impact, which have not been considered by researchers yet. Summary of the invention

[0005] The present invention provides a life prediction method, system and device based on ultra-capacitor coupled lithium battery energy storage frequency modulation to solve the problem that the current life prediction model does not consider the diversity of life influencing factors and the influence degree of each influencing factor, resulting in low life prediction accuracy.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a life prediction method based on ultra-capacitor coupled lithium battery energy storage, comprising the following steps:

[0008] S1, select multiple groups of input variables according to the on-site conditions, and preprocess the multiple groups of input variables to obtain multiple groups of preprocessed input variables; the preprocessing of the multiple groups of input variables is specifically: padding the ends of the input variables with zeros so that the multiple groups of input variables have the same length; the input variables include internal resistance, voltage, current, temperature, number of decomposition layers, penalty factors and multiple groups of variables set manually.

[0009] S2, based on the preprocessed multiple groups of input variables, calculate the similarity D between any two groups of input variables, perform linear fitting on the input variables with similarity D∈[0.7,1.3], and obtain multiple groups of fitting variables. The specific steps are:

[0010] S201, based on each group of input variables X=[x1,x2,…,x n ]Get the change rate series of each group of input variables n is the length of the input variable.

[0011] S202, select any two groups of input variables, obtain similarity D based on the change rate series corresponding to the two groups of input variables, when D∈[0.7,1.3], the linearization degree of the two groups of input variables is high, and the two groups of input variables are linearly fitted to obtain a set of fitting variables;

[0012] The similarity D is:

[0013]

[0014] Among them, R ij is the value of the j-th position in the rate of change series of the input variables in the i-th group, R kj The value of the j position in the k-th group of input variable change rate series.

[0015] The fitting variable Y is:

[0016] Y=λ1X i +λ2X k

[0017] Among them, λ1 and λ2 are the fitting constant sequences, X i is the input variable for group i, X k Enter the variables for group k.

[0018] S203, repeat step S202 until all input variable combinations are obtained to obtain multiple groups of fitting variables S3, and obtain multiple groups of corrected variables by correcting the effects of the multiple groups of fitting variables. The specific steps are:

[0019] Inputting multiple groups of fitting variables into the prediction model to obtain a first prediction result;

[0020] Inputting other groups of fitting variables except each group of fitting variables itself into the prediction model to obtain a second prediction result;

[0021] The difference between the first prediction result and the second prediction result is defined as the contribution of each group of fitting variables;

[0022] The inertia weight of each group of fitting variables is obtained according to the contribution of each group of fitting variables i ; The inertia weight i for:

[0023]

[0024] Where N is the number of groups of fitting variables; G i The contribution of each group of fitting variables;

[0025] Multiply each set of fitting variables by the corresponding inertia weight to obtain multiple sets of correction variables.

[0026] S4, inputting multiple groups of correction variables into the prediction model to obtain battery life prediction data.

[0027] In a second aspect, the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0028] In a third aspect, the present invention provides a life prediction system based on ultra-capacitor coupled lithium battery energy storage, comprising:

[0029] A preprocessing module is used to select multiple groups of input variables according to the on-site conditions, and preprocess the multiple groups of input variables to obtain multiple groups of preprocessed input variables;

[0030] The fitting module calculates the similarity D between any two sets of input variables based on the preprocessed multiple sets of input variables, performs linear fitting on the input variables with similarity D∈[0.7,1.3], and obtains multiple sets of fitting variables;

[0031] A correction module, which corrects the effects of multiple sets of fitting variables to obtain multiple sets of correction variables;

[0032] The prediction module inputs multiple groups of correction variables into the prediction model for processing to obtain battery life prediction data.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention discloses a life prediction method based on ultra-capacitor coupled lithium battery energy storage. By selecting multiple groups of input variables closely related to the battery life, the diversity of factors affecting the battery life is taken into account; and the present invention performs linear fitting on the input variables, reduces the error introduced by the nonlinear relationship, and improves the prediction accuracy; at the same time, the present invention corrects the influence effect of the variables after linear fitting, and by setting appropriate weights for different influencing factors, it can accurately reflect their actual influence on the battery life, thereby obtaining a more accurate prediction result and improving the battery life prediction accuracy.

[0035] Furthermore, the present invention discloses a life prediction system based on ultra-capacitor coupled lithium battery energy storage. The preprocessing module preprocesses multiple groups of input variables of field conditions to ensure the quality of input data; the fitting module can judge the degree of linearization based on the preprocessed input variables, and perform linear fitting on variables with a high degree of linearization, effectively converting complex nonlinear relationships into easy-to-process linear relationships, thereby improving the stability and accuracy of the prediction model; the correction module corrects the influencing effects of the fitting variables, taking into account various possible interference factors. Through correction, the influence of these factors on battery life can be more accurately reflected, thereby improving the reliability of the prediction results.

[0036] Furthermore, the present invention has wide applicability and is applicable to life prediction of different types of lithium batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 This is a flow chart of a battery life prediction method based on supercapacitor coupled lithium battery energy storage according to the present invention;

[0039] Figure 2 This is a block diagram of the battery life prediction system based on supercapacitor-coupled lithium battery energy storage of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention described and marked in the drawings here can be arranged and designed in various different configurations.

[0041] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

[0042] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0043] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0044] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0045] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0046] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0047] See also Figure 1 The present invention discloses a battery life prediction method based on supercapacitor coupled lithium battery energy storage frequency modulation, comprising the following steps:

[0048] S1, select variables that affect life according to on-site conditions to obtain multiple groups of input variables.

[0049] According to the conditions of the production site, multiple groups of factors affecting the lifespan are selected as input variables; the multiple groups of input variables affecting the battery lifespan are internal resistance, voltage, current, temperature, number of decomposition layers and penalty factor. Other input variables affecting the battery lifespan can also be artificially selected according to the conditions on site, such as the quantified depth of each maintenance, ambient humidity and frequency of use.

[0050] Batteries with small internal resistance have strong ability to discharge with large current, while batteries with large internal resistance have weak ability to discharge with large current. Excessive internal resistance will lead to lower discharge voltage of battery, high self-heating of battery, shorter discharge time, and serious impact on battery performance and life, so internal resistance is selected as input variable. Voltage that is too high or too low will affect battery life. When battery voltage is too low, even short-term discharge will damage the battery; when battery voltage is too high, it is easy to cause deformation of internal structure of battery and a series of irreversible reactions, thus shortening battery life. The magnitude of current density mainly affects the structure of electrode materials inside battery. Changes in these factors will lead to decreased battery performance and shorten battery life. Temperature has a significant impact on battery life, capacity and stability. The chemical reaction rate in battery will decrease under low temperature environment, and the life and capacity of lithium battery will decrease under high temperature environment. VMD method is used for power allocation of supercapacitor coupled lithium battery energy storage. When VMD (variational mode decomposition) is used for hybrid energy storage system allocation, the number of tap layers and penalty factor do not have direct impact on battery life, but they will indirectly affect the performance optimization and power allocation of hybrid energy storage system, thus affecting battery life. Maintenance depth usually refers to the degree of charging and discharging of the battery during its use, which includes shallow discharge and deep discharge. Shallow discharge refers to charging when the battery power is high, while deep discharge refers to charging after the battery power is almost exhausted. Shallow discharge helps to extend the battery life, while deep discharge accelerates the battery life. Excessive humidity will accelerate the aging process of the battery and reduce the battery life. Excessive use frequency will accelerate the aging of the battery and reduce the battery life.

[0051] S2, performs linear fitting on the multiple sets of input variables obtained in S1 to obtain multiple sets of fitting variables, including:

[0052] S201, preprocessing the selected multiple groups of input variables so that they have the same length, thereby obtaining multiple groups of preprocessed input variables;

[0053] After selecting multiple groups of input variables that affect battery life, the length of each group of input variables is calculated; zero padding is performed on input variables that are insufficient in length so that the selected multiple groups of input variables have the same length.

[0054] S202, solving a change rate series for each group of input variables after preprocessing to obtain a change rate series corresponding to each group of input variables;

[0055] The transformation rate series of each group of input variables is: Each group of input variables is: X = [x1, x2, .., x n ]; n is the length of the input variable.

[0056] S203, solving the similarity of the change rate series corresponding to any two groups of input variables in S202 to obtain a similarity D;

[0057] Select any two sets of input variables and calculate their similarity:

[0058]

[0059] Where R ij is the value of the j-th position in the rate of change series of the input variables in the i-th group, R kj The value of the position of the k-th group of input variable change rate series;

[0060] S204, judging the linearization degree of the two groups of input variables by the value of the similarity D, and performing linear fitting on the two groups of input variables with high linearization degree to obtain multiple groups of fitting variables;

[0061] When the similarity D∈[0.7, 1.3], it is considered that the linearization degree of these two sets of input variables is high, and the linear fitting of these two sets of input variables is performed to generate the fitting variable Y;

[0062] Y j =λ 1j X ij +λ 2j X kj

[0063] Then, Y=[Y1,Y2,..,Y n ]

[0064] where Y j is the value of the fitting variable j, λ 1j is the fitting constant for the jth position of the input variable in the i-th group, λ 2j is the fitting constant for the kth group of input variables at position j, X ij The value of the j position of the input variable for the i-th group, X kj is the value of the j position of the input variable for the kth group; the fitting constant λ 1j and λ 2j for:

[0065]

[0066] The above method is used to linearly fit any two groups of input variables with a high degree of linearization to obtain multiple groups of fitting variables.

[0067] S3, performing effect correction on the multiple groups of fitting variables obtained in S2 to obtain multiple groups of corrected variables, including:

[0068] Input the multiple groups of fitting variables obtained in S2 into the prediction model for processing to obtain the first prediction result J;

[0069] Inputting other groups of fitting variables except each group of fitting variables itself into the prediction model for processing to obtain a second prediction result J';

[0070] The contribution of each group of fitting variables is calculated based on the first prediction result and the second prediction result: G i =|J-J'|;

[0071] According to the contribution G of each group of fitting variables i Compute the inertia weights for each set of fitted variables:

[0072]

[0073] Where N is the number of groups of fitting variables;

[0074] Multiply each set of fitting variables by the corresponding inertia weight to obtain N sets of correction variables.

[0075] S4, inputting the multiple groups of modified variables obtained in S3 into the prediction model for processing to obtain the predicted lifespan.

[0076] The prediction models described in the present invention include BP, SVM and GRU; any model that can achieve life prediction is applicable to the present invention.

[0077] The present invention also discloses a battery life prediction system based on supercapacitor coupled lithium battery energy storage, comprising:

[0078] A preprocessing module is used to select multiple groups of input variables according to the on-site conditions, and preprocess the multiple groups of input variables to obtain multiple groups of preprocessed input variables;

[0079] The fitting module calculates the similarity D between any two sets of input variables based on the preprocessed multiple sets of input variables, performs linear fitting on the input variables with similarity D∈[0.7,1.3], and obtains multiple sets of fitting variables;

[0080] A correction module, which corrects the effects of multiple sets of fitting variables to obtain multiple sets of correction variables;

[0081] The prediction module inputs multiple groups of correction variables into the prediction model for processing to obtain battery life prediction data.

[0082] The present invention also discloses an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0083] Example:

[0084] The present invention is described in detail below:

[0085] S1, select internal resistance r, voltage U, current I, temperature t, decomposition layer number k, penalty factor α, each maintenance depth h and ambient humidity hu as input variables:

[0086] U=[U1,U2,…,U n ],r=[r1,r2,…,r n ],….

[0087] The input variables are variables that affect the life span selected based on the on-site conditions of the capacitor-coupled lithium battery energy storage; batteries with small internal resistance have strong ability to use large current discharge, while batteries with large internal resistance have weak large current discharge ability. Excessive internal resistance will lead to a decrease in the battery discharge working voltage, high battery self-heating, shortened discharge time, and serious impact on battery performance and life, so internal resistance is selected as the input variable. Voltage that is too high or too low will affect the battery life. When the battery voltage is too low, even a short-term discharge will damage the battery; when the battery voltage is too high, it is easy to cause deformation of the internal structure of the battery and a series of irreversible reactions, thereby shortening the battery life. The magnitude of the current density mainly affects the structure of the electrode material inside the battery. Changes in these factors will lead to a decrease in battery performance and thus shorten the battery life. Temperature has a significant effect on battery life, capacity and stability. The chemical reaction rate in the battery will decrease in a low temperature environment, and the life of the lithium battery will decrease and the capacity will decrease in a high temperature environment. The VMD method is used for power distribution of supercapacitor coupled lithium battery energy storage. When VMD (variational mode decomposition) is used for hybrid energy storage system distribution, the number of tap layers and penalty factors do not have a direct impact on battery life, but they will indirectly affect the performance optimization and power distribution of the hybrid energy storage system, which may in turn affect the battery life. Maintenance depth usually refers to the degree of charging and discharging of the battery during battery use, which includes shallow discharge and deep discharge. Shallow discharge refers to charging when the battery power is high, while deep discharge refers to charging after the battery power is almost exhausted. Shallow discharge helps to extend the battery life, while deep discharge accelerates the battery life.

[0088] S2, linear fitting is performed on the eight input variables of internal resistance r, voltage U, current I, temperature t, decomposition layer number k, penalty factor α, maintenance depth h and ambient humidity hu to obtain multiple sets of fitting variables, including:

[0089] S201, pre-process the internal resistance r, voltage U, current I, temperature t, decomposition layer number k, penalty factor α, each maintenance depth h and ambient humidity hu, that is, calculate the length of each group of input variables for these eight groups of input variables; then fill the input variables with insufficient length with zeros so that the selected multiple groups of input variables have the same length.

[0090] S202, solving the change rate series for each group of input variables after preprocessing, then the change rate series of the internal resistance r is:

[0091]

[0092] The rate of change of the input variable voltage U is:

[0093]

[0094] According to this method, the series of the change rate of current I, temperature t, decomposition layer number k, penalty factor α, each maintenance depth h and ambient humidity hu are obtained in turn.

[0095] S203, solving the similarity of the change rate series corresponding to any two groups of input variables in S202 to obtain a similarity D;

[0096] The similarity of the input variables internal resistance r and voltage U is:

[0097]

[0098] Where R rj is the value of the rate of change of the internal resistance r, R Uj is the value at the same position of the voltage U change rate series; according to this method, the similarity of any other two groups of input variables is solved;

[0099] S204, judging the linearization degree of the two groups of input variables by the value of the similarity D, and performing linear fitting on the two groups of input variables with high linearization degree to obtain multiple groups of fitting variables;

[0100] When the similarity D∈[0.7, 1.3], it is considered that the linearization degree of these two sets of input variables is high, and the linear fitting of these two sets of input variables is performed to generate the fitting variable Y;

[0101] The fitting variable Y of the input variables internal resistance r and voltage U is:

[0102] Y j =λ 1j r j +λ 2j U j

[0103] Then, Y=[Y1,Y2,..,Y n ]

[0104] where λ 1j is the fitting constant of the input variable internal resistance r at position j, λ 2j is the fitting constant of the input variable voltage U at position j, r j is the value of the input variable internal resistance r at position j, U j is the value of the input variable voltage U at position j; the fitting constant λ 1j and λ 2j for:

[0105]

[0106] The above method is used to linearly fit any two groups of input variables with a high degree of linearization to obtain N groups of fitting variables.

[0107] S3, perform effect correction on the N groups of fitting variables obtained in S2 to obtain N groups of corrected variables, including:

[0108] Input the N groups of fitting variables obtained in S2 into the prediction model for processing to obtain the first prediction result J;

[0109] Inputting the other N-1 groups of fitting variables except each group of fitting variables itself into the prediction model for processing to obtain the second prediction result J';

[0110] The contribution of each group of fitting variables is calculated based on the first prediction result and the second prediction result: G i =|J-J'|;

[0111] According to the contribution G of each group of fitting variables i Compute the inertia weights for each set of fitted variables:

[0112]

[0113] Multiply each set of fitting variables by the corresponding inertia weight to obtain N sets of correction variables.

[0114] S4, inputting the multiple groups of modified variables obtained in S3 into the prediction model for processing to obtain the predicted lifespan.

[0115] The prediction models described in the present invention include BP / SVM / GRU; any model that can achieve life prediction is applicable to the present invention.

[0116] In order to further verify the advantages of the present invention, the present invention selects 800 cycles under frequency modulation instruction 1 and 800 cycles under frequency modulation instruction 2 of a power plant, and uses the prediction method of the present invention and the traditional prediction method to predict the life of lithium batteries respectively, 80% of the data is used for training, and 20% of the data is used for verification; the root mean square error between the two prediction methods and the actual battery life is calculated, and the results are as follows:

[0117] Traditional model (RMSE) The present invention (RMSE) FM command 1 9.2 6.3 FM command 2 4.6 2.2

[0118] like Figure 2 As shown, the present invention discloses a life prediction system based on ultra-capacitor coupled lithium battery energy storage, comprising:

[0119] A preprocessing module is used to select multiple groups of input variables according to the on-site conditions, and preprocess the multiple groups of input variables to obtain multiple groups of preprocessed input variables;

[0120] The fitting module calculates the similarity D between any two sets of input variables based on the preprocessed multiple sets of input variables, performs linear fitting on the input variables with similarity D∈[0.7,1.3], and obtains multiple sets of fitting variables;

[0121] A correction module, which corrects the effects of multiple sets of fitting variables to obtain multiple sets of correction variables;

[0122] The prediction module inputs multiple groups of correction variables into the prediction model for processing to obtain battery life prediction data.

[0123] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A life prediction method based on ultracapacitor coupled lithium battery energy storage, characterized in that: The following steps are involved: S1, selecting multiple groups of input variables according to on-site conditions, and preprocessing the multiple groups of input variables to obtain multiple groups of preprocessed input variables; S2, based on multiple sets of preprocessed input variables, calculates the similarity between any two sets of input variables D , for similarity ∈[0.7,1.3] to perform linear fitting on the input variables and obtain multiple sets of fitting variables; S201, based on each set of input variables Get the change rate series of each group of input variables , n is the length of the input variable; S202, select any two groups of input variables, and obtain similarity based on the change rate series corresponding to the two groups of input variables ,when ∈[0.7,1.3], the linearization degree of the two sets of input variables is high, and the two sets of input variables are linearly fitted to obtain a set of fitting variables; Similarity for: in, For the i The rate of change series of the group input variables j The value of the position, For the k Group input variable change rate series j The value of the position; Fitting variables for: in, and is the fitting constant sequence, For the i Group input variables, For the k Group input variables; S203, repeating step S202 until all input variable combinations are obtained to obtain multiple groups of fitting variables; S3, by correcting the effects of multiple groups of fitting variables, multiple groups of corrected variables are obtained; S4, inputting multiple groups of correction variables into the prediction model to obtain battery life prediction data.

2. The life prediction method based on ultra-capacitor coupled lithium battery energy storage according to claim 1 is characterized in that: In step S1, the preprocessing of multiple groups of input variables is specifically as follows: Pad the end of the input variables with zeros so that multiple groups of input variables have the same length.

3. The life prediction method based on supercapacitor coupled lithium battery energy storage according to claim 2 is characterized in that: The input variables include internal resistance, voltage, current, temperature, number of decomposition layers, penalty factors and multiple groups of variables set manually.

4. The life prediction method based on ultra-capacitor coupled lithium battery energy storage according to claim 1 is characterized in that: Step S3 is specifically as follows: Inputting multiple groups of fitting variables into the prediction model to obtain a first prediction result; Inputting other groups of fitting variables except each group of fitting variables itself into the prediction model to obtain a second prediction result; The difference between the first prediction result and the second prediction result is defined as the contribution of each group of fitting variables; The inertia weight of each group of fitting variables is obtained according to the contribution of each group of fitting variables ; Multiply each set of fitting variables by the corresponding inertia weight to obtain multiple sets of correction variables.

5. The life prediction method based on ultra-capacitor coupled lithium battery energy storage according to claim 4 is characterized in that: The inertia weight for: in, N is the number of groups of fitting variables; The contribution of each set of fitted variables.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

7. A life prediction system based on ultra-capacitor coupled lithium battery energy storage, based on the life prediction method of ultra-capacitor coupled lithium battery energy storage according to claim 1, characterized in that: include: A preprocessing module is used to select multiple groups of input variables according to the on-site conditions, and preprocess the multiple groups of input variables to obtain multiple groups of preprocessed input variables; The fitting module calculates the similarity between any two sets of input variables based on multiple sets of preprocessed input variables. D , for similarity D ∈[0.7,1.3] to perform linear fitting on the input variables and obtain multiple sets of fitting variables; A correction module, which corrects the effects of multiple sets of fitting variables to obtain multiple sets of correction variables; The prediction module inputs multiple groups of correction variables into the prediction model for processing to obtain battery life prediction data.

Images