Power battery state of charge prediction method and system based on multi-intelligent algorithm combination

Abstract

The present invention provides a method and system for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms. A first set of SOC data is obtained by using a lithium battery OCV-SOC test; a moth-to-flame algorithm is integrated with an artificial rabbit optimization algorithm, and an adaptive weight factor is introduced to obtain an improved artificial rabbit optimization algorithm; based on internationally public power battery charge and discharge data, the improved artificial rabbit optimization algorithm is used to optimize the support vector machine algorithm to obtain an improved support vector machine prediction model; the improved prediction model is used to predict the lithium battery OCV-SOC test data to obtain a second set of battery SOC data; the two parts of data are integrated, and the integrated data is trained using the improved support vector machine prediction model to obtain a new battery SOC prediction model; using the prediction model, the battery current, voltage and battery temperature data are acquired in real time, and real-time battery SOC data is predicted. The present invention can improve the accuracy and precision of the power battery state of charge prediction results.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery parameter prediction, and specifically relates to a method and system for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Accurate battery state of charge (SOC) estimation of power batteries (such as lithium batteries) is crucial for the battery management system (BMS) of electric vehicles (EVs). The accuracy of battery SOC prediction has a significant impact on battery charging and discharging efficiency and battery life. Predicted SOC values ​​that are too large or too small may lead to overcharging, overcurrent or power loss, which will reduce battery life and even cause vehicle fire or explosion. Especially when electric vehicles are driving at high speeds, accurately predicted SOC is a key indicator to ensure the safety of drivers and passengers. However, since the SOC of lithium batteries is an internal state of the battery that cannot be directly measured, it is difficult to accurately determine the SOC. In addition, the prediction of battery SOC is also affected by many factors, such as complex working conditions, the accuracy of the battery model, and the accuracy of the detection parameters.

[0004] At present, there are many power battery SOC prediction methods that have been widely used. Commonly used single-cell battery SOC estimation methods include open circuit voltage method, ampere-hour integration method, Kalman filter algorithm, etc. Among them, the open circuit voltage method obtains the open circuit voltage value under different SOC through experiments, and then obtains the corresponding relationship between open circuit voltage (OCV) and SOC. During use, the SOC value at this time can be obtained by measuring the static voltage. The disadvantage is that it can only be used in an offline state, and the detected parameters have certain measurement errors, which makes the SOC estimation accuracy not high. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a method and system for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms. The present invention can improve the accuracy and precision of the prediction results of the state of charge of the power battery.

[0006] According to some embodiments, the present invention adopts the following technical solutions:

[0007] A method for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms comprises the following steps:

[0008] The lithium battery open circuit voltage and state of charge (OCV-SOC) test is used to obtain the relationship between battery voltage, current and temperature and battery SOC, and then obtain the first set of battery state of charge data SOC a ;

[0009] The spiral flight mode of moths in the moth-to-fire algorithm is combined with the detour foraging and random hiding of artificial rabbits in the artificial rabbit optimization algorithm, and an adaptive weight factor is introduced to obtain the improved artificial rabbit optimization algorithm (IARO).

[0010] Based on the data from NASA, an international public battery charging and discharging data collection platform, the improved artificial rabbit optimization algorithm is used to optimize the support vector machine algorithm (SVM) to obtain the improved support vector machine (IARO_SVM) prediction model; the power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the IARO_SVM prediction model to predict the second group of battery state of charge data SOC s ;

[0011] Select the first set of battery state of charge data SOC a 50% and the second set of battery state of charge data SOC s 50% is integrated to obtain the battery SOC data SOC m , based on the improved support vector machine prediction model, the integrated samples are trained to obtain a new improved support vector machine (NIARO_SVM) lithium battery SOC prediction model;

[0012] The prediction model is used to predict the real-time battery state of charge data based on the current, voltage and battery temperature data of the power battery acquired in real time.

[0013] A power battery state of charge prediction system based on a combination of multiple intelligent algorithms, comprising:

[0014] The lithium battery OCV-SOC test module is configured to use the lithium battery OCV-SOC test to obtain the relationship between the battery voltage, current and temperature and the battery SOC, and then obtain the first set of battery state of charge data SOC a ;

[0015] The improvement module is configured to merge the spiral flight mode of the moth in the moth-to-fire algorithm with the detour foraging and random hiding of the artificial rabbit in the artificial rabbit optimization algorithm, and introduce an adaptive weight factor to obtain an improved artificial rabbit optimization algorithm;

[0016] The optimization module is configured to optimize the support vector machine algorithm based on the data from the NASA data set of international public battery charging and discharging data, and use the improved artificial rabbit optimization algorithm to obtain the improved support vector machine prediction model; the power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the improved support vector machine prediction model to predict the second set of battery state of charge data SOC s ;

[0017] The integration module is configured to select a first set of battery state of charge data SOC a A portion of the second group of battery state of charge data SOC s A part of the battery SOC data SOC is integrated to obtain the integrated battery SOC m , based on the improved support vector machine prediction model, the integrated samples are trained to obtain a new improved support vector machine lithium battery SOC prediction model;

[0018] The prediction module is configured to use the prediction model to predict the real-time battery state of charge data based on the current, voltage and battery temperature data of the power battery acquired in real time.

[0019] A computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the steps in the above method are completed.

[0020] An electronic device comprises a memory and a processor, and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps in the above method are completed.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The present invention predicts the state of charge of a power battery based on a combination of multiple intelligent algorithms. The spiral flight mode of the moth-to-flame algorithm is integrated with the detour foraging and random hiding of an artificial rabbit, which can improve the local search capability. An adaptive weight factor is introduced to effectively balance the search capability of the algorithm at different stages, enhance the capability of local development, and find the target solution more quickly in the middle and late stages of the algorithm iteration. This can promote a smooth transition between the global and local searches of the algorithm, improve the optimization performance and stability of the algorithm, and ensure the timeliness and accuracy of the calculation.

[0023] The present invention is based on the data in the NASA data set, an international public battery charge and discharge database, and uses the above-mentioned fused multi-intelligent algorithm to optimize the penalty factor C and variance g of the radial basis kernel function of the support vector machine algorithm to obtain an improved support vector machine algorithm. The data in the international public battery database is highly accurate, which can improve the accuracy of the support vector machine battery SOC prediction model and ensure the accuracy of the battery SOC estimation.

[0024] The present invention integrates the battery charge data obtained from the lithium battery OCV-SOC test and the battery state of charge data predicted by the improved support vector machine prediction model to obtain more comprehensive and accurate battery state of charge information, and the test data and the algorithm calculation results complement and correct each other, which helps to eliminate possible errors and deviations and improve the quality and credibility of the data. The integrated data samples are then trained with the improved support vector machine algorithm to obtain a new improved support vector machine lithium battery SOC prediction model.

[0025] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0027] Figure 1 The present invention is a schematic diagram of a method flow of an embodiment. DETAILED DESCRIPTION

[0028] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0029] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0030] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0031] Embodiment 1

[0032] A power battery state of charge prediction method based on a combination of multiple intelligent algorithms, such as Figure 1 As shown, the following steps are included:

[0033] The lithium battery OCV-SOC test is used to obtain the relationship between battery voltage, current and temperature and battery SOC, and then the first set of battery state of charge data SOC is obtained. a ;

[0034] The spiral flight mode of moths in the moth-to-fire algorithm is integrated with the detour foraging and random hiding of artificial rabbits in the artificial rabbit optimization algorithm, and an adaptive weight factor is introduced to obtain an improved artificial rabbit optimization algorithm.

[0035] Based on the data from NASA data set, an international public battery charging and discharging data, the improved artificial rabbit optimization algorithm is used to optimize the support vector machine algorithm to obtain an improved support vector machine prediction model; the power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the improved support vector machine prediction model to predict the second set of battery state of charge data SOC s ;

[0036] Select the first set of battery state of charge data SOC a 50% and the second set of battery state of charge data SOC s 50% is integrated to obtain the battery SOC data SOC m , based on the improved support vector machine prediction model, the integrated samples are trained to obtain a new improved support vector machine lithium battery SOC prediction model;

[0037] The prediction model is used to predict the real-time battery state of charge data based on the current, voltage and battery temperature data of the power battery acquired in real time.

[0038] The specific contents of each process of the above steps are introduced below.

[0039] The lithium battery OCV-SOC test is used to obtain the relationship between battery voltage, current and temperature and battery SOC, and then the first set of battery state of charge data SOC is obtained. a , the specific process is as follows:

[0040] Under different ambient temperatures, the battery is discharged at different discharge rates for a certain period of time, left to stand, and the battery open circuit voltage is measured, and then the discharge and standing are repeated, and the battery open circuit voltage is measured. The corresponding relationship between the battery voltage, current and temperature and the battery SOC is obtained.

[0041] The spiral flight mode of the moth in the moth-to-flame algorithm is integrated with the detour foraging and random hiding of the artificial rabbit in the artificial rabbit optimization algorithm, and an adaptive weight factor is introduced to obtain the improved artificial rabbit optimization algorithm. The artificial rabbit optimization algorithm is first introduced.

[0042] The Artificial Rabbit Optimization (ARO) algorithm has the advantages of adaptability, global search capability, high efficiency, robustness, parallelization capability, and easy understanding and implementation. These advantages make the ARO algorithm have a wide range of application potential in various optimization problems. The Artificial Rabbit Optimization algorithm uses the foraging and hiding strategies of real rabbits. The algorithm switches between the two strategies through energy contraction, and includes the following three behaviors.

[0043] 1) Take a detour to find food

[0044] When rabbits forage for food, they always look for grass far away and ignore nearby grass. They only eat grass randomly in other areas instead of eating grass in their own area. This foraging behavior is called detour foraging. Its mathematical model is as follows:

[0045] x i (t+1)=x j (t)+R·(x i (t)-x j (t))+round(0.5×(0.05+r1))·n1,

[0046] i, j = 1, ... n and j ≠ i;

[0047] R = L·c;

[0048]

[0049]

[0050] k=1,…,d;

[0051] l = 1, ..., [r3·d];

[0052] g = randperm(d);

[0053] Where: x i (t+1) is the candidate position of the i-th rabbit in the t+1-th iteration; x i (t) is the position of the i-th rabbit in the t-th iteration, x j (t) is the position of the jth rabbit in the tth iteration; R represents the running operator; round() returns the rounded value; c(k) is the mapping vector; randperm(d) represents a random integer between 1 and d; L is the running length; n1 is a random number that follows the standard normal distribution; r1, r2, and r3 are random numbers between (0, 1); d is the individual dimension of the artificial rabbit; and T is the maximum number of iterations.

[0054] 2) Random hiding

[0055] In each iteration, a rabbit always generates a hole along the dimension of the search space, and always randomly selects one hole from all the holes to hide in to reduce the probability of being preyed on. When rabbits encounter danger, they randomly select a hole to hide. This random hiding strategy model is as follows:

[0056] x i (t+1)=x i (t)+R·(r4·bi，j (t)-x i (t));

[0057] i=1,…,n

[0058] b i，j (t) = x i (t)+H·g r ·x i (t);

[0059]

[0060]

[0061] Where: b i,j (t) is the jth hole generated by the i-th rabbit in the t-th iteration; H is the hidden parameter; n2 is a random number from the standard normal distribution; r4, r5 are random numbers between (0, 1); g r (k) is the mapping vector. After implementing the detour foraging or random hiding strategy, the rabbit's position is updated as:

[0062]

[0063] Where: f(x) is the fitness function used to optimize the problem.

[0064] 3) Energy reduction

[0065] The switch between the detour foraging and random hiding strategies is achieved through energy reduction. Over time, the rabbit's energy will gradually decrease. The energy factor in ARO is defined as follows:

[0066]

[0067] Where r is a random number between (0, 1). As time increases, the energy factor A(t) tends to 0. When A(t)>1, the individuals in the population implement a detour foraging strategy; otherwise, the individuals in the population implement a random hiding strategy.

[0068] Although ARO outperforms many classic nature-inspired algorithms, it still has some shortcomings, such as lack of population diversity and easy to fall into local optimality.

[0069] In order to solve the above shortcomings of the artificial rabbit optimization algorithm, this embodiment makes the following improvements:

[0070] The Moth-to-Fire Algorithm (MFO) is an efficient bionic intelligent optimization algorithm. It is a bionic optimization algorithm proposed by observing and imitating the behavior of moths flying around light sources. Moths fly around the flame in a spiral, constantly searching for better candidate solutions near the flame and updating the flame position. The spiral flight method can improve the local search capability. The update formula is shown in the following formula.

[0071] S(M i , F j )=|F j -M i |·e bθ ·cos(2πθ)+F j

[0072] Among them, M i is the i-th moth; F j is the jth flame; b is the constant of the spiral function; θ is a random number in [-1, 1];

[0073] Based on the artificial rabbit algorithm, this embodiment integrates the spiral flight mode of moths in MFO with the detour foraging and random hiding of artificial rabbits. The detour foraging formula after integration is as follows:

[0074] x i (t+1)=x new (t)+R·(x i (t)-x new (t))+round(0.5·(0.05+r1))·n1;

[0075] x new (t)=|x great -x i |·e bθ ·cos(2πθ)+x great ;

[0076] The fused random hidden formula is as follows:

[0077] x i (t+1)=x new (t)+R·(r4·b i，j (t)-x new (t));

[0078] where x new (t) is the position of the artificial rabbit after the moth spiral flight is updated, x great is the optimal artificial rabbit position.

[0079] From the above formula, we can know that the value of θ can show the distance between the ith artificial rabbit and the optimal artificial rabbit. When θ is -1, the ith artificial rabbit is closest to the optimal artificial rabbit; when θ is 1, the ith artificial rabbit is farthest from the optimal artificial rabbit, that is, the smaller the value of θ is, the closer the distance between the ith artificial rabbit and the optimal artificial rabbit is. By continuously adjusting the value of θ, the artificial rabbit can traverse the entire search space. In order to make the algorithm have a strong optimization ability in the early stage and a strong local development ability in the later stage of iteration, it is necessary to improve the value strategy of θ, and stipulate that θ is randomly valued in [u, 1] (u is a variable that decreases linearly from -1 to -2 with the number of iterations).

[0080] Based on the above, this embodiment further introduces an adaptive weight factor to effectively balance the search capabilities of the algorithm at different stages, as follows:

[0081]

[0082] Where: t is the current iteration number; T is the maximum iteration number.

[0083] By introducing the adaptive weight factor, the position update formula of the detour foraging and hidden random mode search route in ARO becomes as follows:

[0084] Detour foraging route update:

[0085] x i (t+1)=x new (t)·ω+R·(x i (t)-x new (t))+round(0.5·(0.05+r1))·n1;

[0086] x new (t)=|x great -x i |·e bθ ·cos(2πθ)+x great ;

[0087] Random hidden route update:

[0088] x i (t+1)=x new (t)·ω+R·(r4·b i,j (t)-x new (t));

[0089] As the number of iterations increases, the value of ω increases nonlinearly. In the early stage of the algorithm iteration, the value of the adaptive weight factor is small, the location of the food is fuzzy, and it has little effect on the update of detour foraging and random hiding in the original artificial rabbit algorithm; in the middle and late stages of the algorithm iteration, the food gradually becomes clear. At this time, the value of the adaptive weight factor gradually increases with the increase in the number of iterations. The adaptive weight factor enhances the ability of local development, allowing the rabbit to find the target solution faster in the middle and late stages of the algorithm iteration. Adding an adaptive weight factor can promote a smooth transition between the global and local searches of the algorithm and improve the algorithm's optimization performance and stability.

[0090] The above improvement process can obtain the improved artificial rabbit algorithm (IARO).

[0091] In this embodiment, the SVM algorithm can be trained based on the data in the NASA data set, an international public battery charge and discharge database. The international public battery database is open to the public, and it is easy for technicians in this field to select any or several database combinations, which are not exhaustive here.

[0092] Since the battery SOC is related to the battery current, battery voltage, and battery temperature data, the SOC is regressed and predicted with the support vector machine algorithm based on the battery current, voltage, and temperature. However, in order to ensure the accuracy of the prediction, the support vector machine algorithm is improved as follows:

[0093] In this embodiment, the battery current, battery voltage, and battery temperature in the international public NASA data set are used as SVM inputs, and the battery SOC is used as output. The radial basis kernel function is selected as the kernel function of the SVM, and the formula is:

[0094] K(X i , X j )=exp(-G g ||X i -X j || 2 ), G g >0

[0095] In the formula, X i and X j These two parameters represent two sample points (feature vectors) in the data set, each of which consists of battery current, battery voltage, and battery temperature; G g is the variance of the radial basis kernel function.

[0096] In order to improve the performance of SVM, the improved IARO algorithm is used to improve SVM.

[0097] The penalty factor C and variance G of the radial basis kernel function of the SVM algorithm are adjusted using the IARO algorithm. gTo optimize, the battery SOC mean square error mse of the SVM algorithm regression and the square correlation coefficient R of the regression 2 As the fitness value of the IARO algorithm, the position corresponding to the best fitness value is the optimal position of the rabbit, and the improved support vector machine prediction model is obtained.

[0098] Select the mean square error mse and the squared correlation coefficient R of the regression 2 To evaluate the accuracy of training, the closer the value of mse is to 0, R 2 The closer the value is to 1, the more accurate the model is and the stronger the generalization ability of the algorithm is. 2 The calculation formula is as follows:

[0099]

[0100]

[0101] In the formula, n is the number of samples, s is i Battery SOC data from the international public NASA dataset. To predict data, It is the mean value of battery SOC data of the international public NASA data set.

[0102] In this embodiment, 70% of the samples of voltage, current, temperature and battery SOC data in the international public battery charge and discharge database NASA are randomly selected as the training set, and 30% of the samples are used as the test set. Of course, in other embodiments, other proportions or methods can be used to form the training set and the test set.

[0103] Similarly, in other embodiments, other evaluation indicators may be used to evaluate the performance of the optimized model, and the training is stopped when the test performance meets the requirements.

[0104] In this embodiment, when optimizing the improved support vector machine algorithm, the battery SOC mean square error and the square correlation coefficient of regression are used to evaluate the accuracy of training. When the mean square error is close to 0 and the square correlation coefficient is close to 1 and tends to be stable, the optimization is terminated.

[0105] Get the best C and G g Finally, the current, voltage and battery temperature data obtained from the lithium battery OCV-SOC test are used as the input of the improved support vector machine prediction model, and the predicted battery SOC is used as the output. The data is sorted to obtain the battery SOC data predicted by the improved support vector machine. This data is the second set of battery SOC data SOC s , select the lithium battery OCV-SOC test to obtain the battery SOC data SOC a 50% and the battery SOC data SOC predicted by the improved support vector machines 50%, the two parts of data are integrated to obtain the new battery SOC data SOC m The integration formula is as follows:

[0106] SOC m =50% SOC a +50% SOC s ;

[0107] Integrated battery data SOC based on improved support vector machine m After training, a new improved support vector machine lithium battery SOC prediction model is obtained.

[0108] New battery current, battery voltage and battery temperature data are collected in real time, and the data is used as input to a new improved support vector machine battery SOC prediction model, and the output is real-time predicted battery SOC data.

[0109] Embodiment 2

[0110] A power battery state of charge prediction system based on a combination of multiple intelligent algorithms, comprising:

[0111] The lithium battery OCV-SOC test module is configured to use the lithium battery OCV-SOC test to obtain the relationship between the battery voltage, current and temperature and the battery SOC, and then obtain the first set of battery state of charge data SOC a ;

[0112] The improvement module is configured to merge the spiral flight mode of the moth in the moth-to-fire algorithm with the detour foraging and random hiding of the artificial rabbit in the artificial rabbit optimization algorithm, and introduce an adaptive weight factor to obtain an improved artificial rabbit optimization algorithm;

[0113] The optimization module is configured to optimize the support vector machine algorithm based on the data from the NASA data set, an international public battery charging and discharging data, using the improved artificial rabbit optimization algorithm to obtain an improved support vector machine prediction model. The power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the improved support vector machine prediction model to predict the second set of battery state of charge data SOC s ;

[0114] The integration module is configured to select a first set of battery state of charge data SOC a 50% and the second set of battery state of charge data SOC s 50% is integrated to obtain the integrated battery SOC data SOC m , based on the improved support vector machine prediction model, the integrated samples are trained to obtain a new improved support vector machine lithium battery SOC prediction model; the integration formula is as follows:

[0115] SOC m =50% SOC a +50% SOC s ;

[0116] The prediction module is configured to use the prediction model to predict the real-time battery state of charge data based on the current, voltage and battery temperature data of the power battery acquired in real time.

[0117] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0118] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0119] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0121] The above description is only a preferred embodiment of the present invention and is 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 by those skilled in the art within the spirit and principle of the present invention without creative labor shall be included in the protection scope of the present invention.

Claims

1. A method for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms, characterized in that: The following steps are involved: The lithium battery OCV-SOC test is used to obtain the relationship between battery voltage, current and temperature and battery SOC, and then the first set of battery state of charge data SOC is obtained. a ; The spiral flight mode of moths in the moth-to-fire algorithm is integrated with the detour foraging and random hiding of artificial rabbits in the artificial rabbit optimization algorithm, and an adaptive weight factor is introduced to obtain an improved artificial rabbit optimization algorithm. Based on the data from NASA data set, an international public battery charging and discharging data, the improved artificial rabbit optimization algorithm is used to optimize the support vector machine algorithm to obtain an improved support vector machine prediction model; the power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the improved support vector machine prediction model to predict the second set of battery state of charge data SOC s ; Select the first set of battery state of charge data SOC a A portion of the second group of battery state of charge data SOC s A part of the battery SOC data SOC is integrated to obtain the integrated battery SOC m , based on the improved support vector machine prediction model, the integrated samples are trained to obtain a new improved support vector machine lithium battery SOC prediction model; Using the new improved support vector machine lithium battery SOC prediction model, based on the current, voltage and battery temperature data of the power battery acquired in real time, the real-time battery SOC data is predicted; The specific process of integrating the spiral flight mode of moths in the moth-to-fire algorithm with the detour foraging and random hiding of artificial rabbits in the artificial rabbit optimization algorithm includes: The update formula for the spiral flight mode of the moth in the moth-to-fire algorithm is: in, For the Only moths; For the a flame; is the constant of the spiral function; A random number in the range [-1, 1]; The artificial rabbits after fusion take detours to forage for food: ； ； ； ； ； in, For the In the iteration Rabbit candidate position; For the In the iteration The position of the rabbit, For the In the iteration The position of the rabbit; represents the running operator; is the mapping vector; Indicates from 1 to A random integer between ; for the running length; is a random number that follows a standard normal distribution; is a random number between (0, 1), The position of the artificial rabbit after using the moth spiral flight update, is the optimal artificial rabbit position, is the maximum number of iterations, To return the rounded value; The fused artificial rabbit is randomly hidden as follows: ； ； ； ； in, For the In the iteration The first rabbit produced a cave; is a hidden parameter; is a random number from a standard normal distribution; , is a random number between (0, 1); is the mapping vector; The introduced adaptive weight factor is: in: is the current iteration number; is the maximum number of iterations; After introducing the adaptive weight factor, the position update formula of the detour foraging is: The random hidden position update formula is: As the number of selection generations increases, its value increases nonlinearly; Thus, an improved artificial rabbit optimization algorithm is obtained.

2. A method for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms as claimed in claim 1, characterized in that The lithium battery OCV-SOC test is used to obtain the relationship between battery voltage, current and temperature and battery SOC, and then the first set of battery state of charge data SOC is obtained. a .

3. The method for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms as claimed in claim 1, characterized in that: Based on the data from NASA data set, an international public battery charging and discharging data, the penalty factor and variance of the radial basis kernel function of the support vector machine algorithm are optimized using the improved artificial rabbit optimization algorithm. The optimization goal is to obtain the optimal penalty factor and variance values ​​so that the root mean square error of the regression is minimized. When optimizing the improved support vector machine algorithm, the battery SOC mean square error and the square correlation coefficient of regression are used to evaluate the accuracy of the training. When the mean square error is close to 0 and the square correlation coefficient is close to 1 and tends to be stable, the optimization is terminated, thereby obtaining an improved support vector machine prediction model. The power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the improved support vector machine prediction model to predict the second set of battery state of charge data SOC s .

4. The method for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms as claimed in claim 1, characterized in that: Select the first set of battery state of charge data SOC a 50% data of the battery, and the second set of battery state of charge data SOC s 50% of the data is integrated to obtain the battery SOC data SOC m ; The improved support vector machine algorithm is trained using the integrated data to obtain a new improved support vector machine lithium battery SOC prediction model.

5. A method for predicting the state of charge of a power battery based on a combination of multiple intelligent algorithms as claimed in claim 4, characterized in that: The new improved support vector machine lithium battery SOC prediction model is used to predict the real-time battery SOC data based on the current, voltage and battery temperature data of the power battery acquired in real time.

6. A power battery state of charge prediction system based on a combination of multiple intelligent algorithms, characterized by comprising: The lithium battery OCV-SOC test module is configured to use the lithium battery OCV-SOC test to obtain the relationship between the battery voltage, current and temperature and the battery SOC, and then obtain the first set of battery state of charge data SOC a ; The improvement module is configured to merge the spiral flight mode of the moth in the moth-to-fire algorithm with the detour foraging and random hiding of the artificial rabbit in the artificial rabbit optimization algorithm, and introduce an adaptive weight factor to obtain an improved artificial rabbit optimization algorithm; The optimization module is configured to optimize the support vector machine algorithm based on the data from the NASA data set of international public battery charging and discharging data, and obtain the improved support vector machine prediction model by using the improved artificial rabbit optimization algorithm; the power battery current, voltage and temperature data obtained from the lithium battery OCV-SOC test are introduced into the improved support vector machine prediction model to predict the second set of battery state of charge data SOC s ; The integration module is configured to select a first set of battery state of charge data SOC a A portion of the second group of battery state of charge data SOC s A part of the battery SOC data SOC is integrated to obtain the integrated battery SOC m , based on the improved support vector machine prediction model, the integrated samples are trained to obtain a new improved support vector machine lithium battery SOC prediction model; A prediction module is configured to use the new improved support vector machine lithium battery SOC prediction model to predict the real-time battery state of charge data based on the current, voltage and battery temperature data of the power battery acquired in real time; The specific process of integrating the spiral flight mode of moths in the moth-to-fire algorithm with the detour foraging and random hiding of artificial rabbits in the artificial rabbit optimization algorithm includes: The update formula for the spiral flight mode of the moth in the moth-to-fire algorithm is: in, For the Only moths; For the a flame; is the constant of the spiral function; A random number in the range [-1, 1]; The artificial rabbits after fusion take detours to forage for food: ； ； ； ； ； in, For the In the iteration Rabbit candidate position; For the In the iteration The position of the rabbit, For the In the iteration The position of the rabbit; represents the running operator; is the mapping vector; Indicates from 1 to A random integer between ; for the running length; is a random number that follows a standard normal distribution; is a random number between (0, 1), The position of the artificial rabbit after using the moth spiral flight update, is the optimal artificial rabbit position, is the maximum number of iterations, To return the rounded value; The fused artificial rabbit is randomly hidden as follows: ； ； ； ； in, For the In the iteration The first rabbit produced a cave; is a hidden parameter; is a random number from a standard normal distribution; , is a random number between (0, 1); is the mapping vector; The introduced adaptive weight factor is: in: is the current iteration number; is the maximum number of iterations; After introducing the adaptive weight factor, the position update formula of the detour foraging is: The random hidden position update formula is: As the number of selection generations increases, its value increases nonlinearly; Thus, an improved artificial rabbit optimization algorithm is obtained.

7. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the steps of the method according to any one of claims 1 to 5.

8. An electronic device, characterized in that: The method comprises a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the steps in the method according to any one of claims 1 to 5 are completed.

Images