CN113325317A - Power battery fault diagnosis method and system based on improved RBF neural network - Google Patents

Power battery fault diagnosis method and system based on improved RBF neural network Download PDF

Info

Publication number
CN113325317A
CN113325317A CN202110774153.1A CN202110774153A CN113325317A CN 113325317 A CN113325317 A CN 113325317A CN 202110774153 A CN202110774153 A CN 202110774153A CN 113325317 A CN113325317 A CN 113325317A
Authority
CN
China
Prior art keywords
neural network
power battery
rbf neural
fault diagnosis
center
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110774153.1A
Other languages
Chinese (zh)
Inventor
张承慧
刘振宇
李岩
商云龙
段彬
崔纳新
张奇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University
Original Assignee
Shandong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong University filed Critical Shandong University
Priority to CN202110774153.1A priority Critical patent/CN113325317A/en
Publication of CN113325317A publication Critical patent/CN113325317A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/378—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] specially adapted for the type of battery or accumulator

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Secondary Cells (AREA)

Abstract

The utility model provides a power battery fault diagnosis method and system based on an improved RBF neural network, which obtains the parameter data of the power battery; obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model; in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm; the method improves the precision of fault diagnosis of the power battery, is beneficial to improving the use safety of the power battery and prolongs the service life of the power battery.

Description

Power battery fault diagnosis method and system based on improved RBF neural network
Technical Field
The disclosure relates to the technical field of power battery fault diagnosis, in particular to a power battery fault diagnosis method and system based on an improved RBF neural network.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The power battery is used as the most important part of the pure electric vehicle, the performance of the pure electric vehicle is directly determined by the operating state of the power battery, when the power battery breaks down, the operating state of the whole electric vehicle can be directly caused to be in a problem, the power battery can seriously cause short circuit, fire and even explosion accidents, and great threat is formed to the driving safety of the electric vehicle. Therefore, the battery fault is diagnosed and checked safely, efficiently and in real time, and the method has great significance for prolonging the service life of the battery and improving the driving safety of the electric automobile.
The power battery is a nonlinear real variable system, and the system performance is influenced by various parameter changes such as voltage, current, temperature, battery internal resistance and the like. At present, the power battery fault diagnosis method can be roughly divided into two categories: the fault diagnosis method based on the battery model needs to establish an accurate and reliable battery model, carries out fault diagnosis by comparing an actual measurement value with a predicted value of the battery model, and is difficult to be suitable for real-time online diagnosis due to high model complexity. The battery-free model fault diagnosis method does not need to establish a complex battery model, and carries out fault diagnosis on line in real time by using an expert system or a series of algorithms such as data analysis, data mining and the like according to battery sampling data. The artificial neural network is a nonlinear and self-adaptive information processing system formed by interconnection of a large number of processing units, has self-adaptive, self-organizing and self-learning capabilities, and can effectively process a multi-input nonlinear real-variant system. Therefore, artificial neural networks are often used in the design of fault diagnosis for power cells.
The commonly used artificial neural network structure mainly includes a Back Propagation (BP) network and a Radial Basis Function (RBF) network, and the difference between them is that the activation Function used by the hidden layer node is different. The hidden layer of the BP neural network adopts an S-shaped basis function, the hidden layer of the BP neural network takes a non-zero value in an infinite area in an input space, and the problems of low calculation speed, easy occurrence of local optimal solution and the like are caused by a plurality of neural units and weight initial value data, complex model training and the like. The hidden layer of the RBF neural network adopts a Gaussian basis function, takes a nonzero value only in a small area in an input space, and locally generates response to an input sample, so that the RBF neural network has local approximation capability, can obtain better precision and running speed when solving the problem of parameter identification, and does not have the condition of training failure caused by falling into a local minimum value.
To build an RBF neural network, three important parameters in the network are determined: the central vector of the basis function, the expansion coefficient of the hidden layer unit and the connection weight between the hidden layer neuron and the output layer neuron. These three parameters will have a great influence on the learning effect of the neural network: the central vector ensures that the input signal of the neural network is in the effective range of the Gaussian function, the expansion coefficient ensures the effective mapping of the neural network, and the connection weight is a key variable for determining whether the RBF neural network can realize accurate approximation.
The inventor finds that in the traditional RBF neural network, no good solution is provided for determining the appropriate basis function center vector and expansion coefficient, so that the accuracy of the final power battery fault diagnosis result is low.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a power battery fault diagnosis method and system based on an improved RBF neural network.
In order to achieve the purpose, the following technical scheme is adopted in the disclosure:
the first aspect of the disclosure provides a power battery fault diagnosis method based on an improved RBF neural network.
A power battery fault diagnosis method based on an improved RBF neural network comprises the following processes:
acquiring parameter data of a power battery;
obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model;
in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm.
Further, the parametric data includes at least one or more of a battery voltage, a battery temperature, a SOC (State of charge), an internal resistance of the battery, and a capacity.
Further, the center of the basis function is determined through a subtractive clustering algorithm, and the method comprises the following processes:
regarding a plurality of input sample data, taking the sample data as a candidate set of a clustering center, and calculating a density index of each data point;
selecting a data point with the highest density index as a first clustering center, and revising the density indexes of other data points by taking the data point as a center;
judging whether the termination criterion of the subtractive clustering algorithm is met;
if the termination criterion of the subtractive clustering algorithm is not met, selecting the point with the highest density index at the moment as a second clustering center, and correcting the density indexes of other data points again by taking the data point as the center;
and repeating the process continuously until the termination criterion of the subtractive clustering algorithm is met, and taking the final clustering center as the center of the basis function.
Furthermore, according to the obtained cluster center, the average distance between the cluster center and the data points adjacent to the cluster center is taken as an expansion coefficient.
Further, the termination criterion of the subtractive clustering algorithm is as follows: and the ratio of the highest density index of all the current data points to the density index of the cluster center is smaller than a preset constant.
Furthermore, after the cluster center is reselected, the density indexes of other data points are revised, and the revision formula is as follows:
Figure BDA0003153784770000041
wherein the content of the first and second substances,
Figure BDA0003153784770000043
as a new density index of the cluster center, xiFor the ith data point, the data point,
Figure BDA0003153784770000044
data points, σ, represented by the cluster centersbIs a preset constant.
Further, the density index for each data point includes:
Figure BDA0003153784770000042
where M is the total number of data points, xiIs the ith data point, xjIs the jth data point, σaIs a preset constant.
The second aspect of the disclosure provides a power battery fault diagnosis system based on an improved RBF neural network.
A power battery fault diagnosis system based on an improved RBF neural network comprises:
a data acquisition module configured to: acquiring parameter data of a power battery;
a fault diagnosis module configured to: obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model;
in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm.
A third aspect of the present disclosure provides a computer-readable storage medium, on which a program is stored, which when executed by a processor, implements the steps in the improved RBF neural network-based power battery fault diagnosis method according to the first aspect of the present disclosure.
A fourth aspect of the present disclosure provides an electronic device, which includes a memory, a processor, and a program stored in the memory and executable on the processor, and the processor executes the program to implement the steps in the method for diagnosing a fault of a power battery based on an improved RBF neural network according to the first aspect of the present disclosure.
Compared with the prior art, the beneficial effect of this disclosure is:
1. according to the power battery fault diagnosis method, the system, the computer readable storage medium or the electronic equipment, the center and the expansion coefficient of the basis function in the RBF neural network are determined through the subtractive clustering algorithm, and the power battery fault diagnosis precision is improved.
2. The power battery fault diagnosis method, the power battery fault diagnosis system, the computer readable storage medium or the electronic equipment can accurately detect the fault type, are favorable for improving the use safety of the power battery and prolonging the service life, and can be applied to actual use and maintenance of the power battery.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
Fig. 1 is a general design diagram of a power battery fault diagnosis scheme provided in embodiment 1 of the present disclosure.
Fig. 2 is a block diagram of an RBF neural network provided in embodiment 1 of the present disclosure.
Fig. 3 is a flow chart of a subtractive clustering algorithm provided in embodiment 1 of the present disclosure.
Fig. 4 is a discharge model diagram of a lithium iron phosphate battery in a fault simulation experiment provided in embodiment 1 of the present disclosure.
Fig. 5 is a diagram of a training process of an improved RBF-based neural network according to embodiment 1 of the present disclosure.
Detailed Description
The present disclosure is further described with reference to the following drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
Example 1:
as shown in fig. 1, an embodiment 1 of the present disclosure provides a power battery fault diagnosis method based on an improved RBF neural network, including the following processes:
acquiring parameter data of a power battery;
obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model;
in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm.
After the fault diagnosis result is given, the battery management system makes certain reactions to the battery fault, including fault information display, alarm, power cut-off and the like.
The RBF neural network is a feedforward neural network, the network topology structure of which is shown in figure 2 and consists of three layers, namely an input layer, a radial basis function hidden layer and an output layer.
The input layer is a plurality of signal source nodes which are only responsible for receiving input signals and do not perform any processing on input data, and the input layer plays a role of connecting with the external environment. The power battery fault diagnosis system designed in the present embodiment is provided with input signals including battery voltage, battery temperature, SOC, battery internal resistance, capacity, etc. by a Battery Management System (BMS).
The input layer passes the input signal to the hidden layer, which uses the radial basis functions therein to effect a non-linear transformation from the input to the hidden layer. The radial basis function of the hidden layer is crucial to the whole neural network, and the selection of the radial basis function must meet the characteristics of simple form, easy analysis, symmetrical attenuation, good smoothness and existence of any reciprocal. Therefore, when the RBF neural network is approximated, the closer the function is to the central point, the more sensitive the output response is, and conversely, the farther the function is from the central point, the less sensitive the output response is, so that the problem of local minimum of the network is solved.
The gaussian function curve is smooth, any order derivative exists, the form is simple, theoretical analysis is facilitated, the radial basis function which is most commonly used in the RBF neural network is formed, and the gaussian function is also selected as the basis function of the hidden layer neuron in the embodiment:
Figure BDA0003153784770000071
in the formula: h isi(x) Is the output of the ith hidden layer cell, ciIs the central vector, σ, of the ith hidden layer celliIs the expansion coefficient of the ith hidden layer unit. The two most important parameters in the gaussian function are the center vector and the expansion coefficient, which need to be determined in the neural network.
The following describes in detail a method for determining the center and the expansion coefficient of a basis function in an RBF neural network by using a subtractive clustering algorithm, where the flow of the subtractive clustering algorithm is shown in fig. 3, and the specific implementation process of the method includes the following steps:
step 1, inputting M N-dimensional sample data xm(m is 1,2, …, m), all of which are used as candidate sets of cluster centers, and the density index D of each data point is calculatedi：
Figure BDA0003153784770000072
In the formula: sigmaaIs a constant greater than zero and defines a field of the data point, and the influence of data points outside the field on the density of the point is almost negligible. Thus, if a data point has a high density indicator, there must be multiple adjacent data points near that point.
Step 2, after the density index of each data point is calculated, selecting a point with the highest density index as a first clustering center, and recording the point as the first clustering center
Figure BDA0003153784770000084
The corresponding density index is
Figure BDA0003153784770000085
Then, taking the data point as the center, revising the density indexes of other data points, wherein the revising formula is as follows:
Figure BDA0003153784770000081
in the formula: sigmabAnd σaSimilarly, it is a constant greater than zero, which defines a field of significant reduction of the density index, avoiding the appearance of very close cluster centers, whose value is generally greater than σa。
Step 3, judging whether the termination criterion of the subtractive clustering algorithm is met:
Figure BDA0003153784770000082
in the formula: dmaxIs the highest density indicator for all data points present and gamma is a small constant. The expression has the significance that the current highest density index is very small compared with the initial density index, the current remaining data points can be ignored as the clustering center, and the clustering is finished when the current remaining data points meet the requirement.
And 4, if the termination criterion of the subtractive clustering algorithm is not met, selecting a point with the highest density index at the moment as a second clustering center, and correcting the density indexes of other data points again by taking the data point as a center, wherein a more general correction formula is as follows:
Figure BDA0003153784770000083
and repeating the complaint process continuously until the termination criterion of the subtractive clustering algorithm is met.
Step 5, after the clustering center is determined, taking the average distance between the clustering center and the data points adjacent to the clustering center as an expansion coefficient sigmai。
The output layer has the function of performing linear transformation on the signal of the hidden layer, namely performing weighted linear summation on the output of the hidden layer.
The output of the output layer can be expressed as:
Figure BDA0003153784770000091
in the formula, yj(x) Is the output signal of the jth neuron of the output layer, N is the number of neurons of the hidden layer, ωijThe weight vector is between the ith neuron of the hidden layer and the jth neuron of the output layer, and in the neural network designed in this embodiment, the output of the output layer is the diagnosis result of various faults.
And finally, the output layer outputs the diagnosis result of the battery fault to the BMS, and the battery management system makes certain reactions to the battery fault, including fault information display, alarming, power supply cut-off and the like.
In order to verify the effectiveness of the embodiment, the embodiment adopts MATLAB simulation to perform a fault simulation experiment on the power battery, and performs fault diagnosis.
Because the actual power battery fault experiment needs irreparable damage to the battery, the cost is high, and certain danger exists, the MATLAB simulation is selected to carry out the fault simulation experiment on the power battery in the embodiment. The MATLAB is provided with a power battery model, the lithium iron phosphate battery model is selected, a corresponding circuit model is built, and the normal state and various fault states in the discharge process of the lithium iron phosphate battery monomer are simulated through setting various parameters of the battery and setting the environmental temperature.
The rated voltage of the lithium iron phosphate battery is 3.3V, the rated capacity is 2.3AH, the rated discharge current is 2.3A, the internal resistance is 0.014 omega, the response time of the battery is set to be 30S, and the initial battery temperature is set to be 20 ℃. The discharge circuit model is shown in fig. 4, the model adopts a constant current source for discharging, the environmental temperature can be set, and three oscilloscopes can test real-time data such as voltage, current, temperature, SOC and the like in the discharge process of the battery. The experiment simulates seven fault states of a normal state and insufficient battery charge, overvoltage, undervoltage, increased internal resistance, reduced battery capacity, overhigh temperature, overlow temperature and the like in the discharge process of the lithium iron phosphate battery, and the data acquisition of each state comprises ten groups of samples, so that eighty groups of sample data are total, and each group of samples comprises five-dimensional data of voltage, temperature, SOC, internal resistance and capacity. Carrying out normalization processing on eighty groups of collected sample data, and then using the sample data for training and testing a neural network, wherein X is a sample input vector matrix, and the data is as follows:
Figure BDA0003153784770000101
the output layer is represented by a binary coding form, and fault codes under 8 states are respectively set: normal (000), low battery charge (001), over-voltage (010), under-voltage (011), increased internal resistance (100), reduced battery capacity (101), over-temperature (110), and under-temperature (111).
Y is the sample output vector matrix, data is as follows:
Figure BDA0003153784770000102
in the experiment, the input of the RBF neural network is five-dimensional signal data, and the output is three-bit binary coding, so that the input layer of the improved RBF neural network in the embodiment has 5 nodes, and the output layer has 3 nodes. The number of hidden layer neurons and the expansion coefficient were determined by a subtractive clustering algorithm, while the mean square error was set to 0.01. 64 groups of samples in 8 groups of states are input into the improved RBF neural network for training, and the rest 16 groups of samples are input into the improved RBF neural network as detection data for testing. The test samples are shown in table 1.
Table 1: test sample data
Figure BDA0003153784770000103
Figure BDA0003153784770000111
The RBF neural network training process based on the subtractive clustering algorithm is shown in FIG. 5, and the training only needs 27 steps to reach the expected target, which shows that the training speed is high and the precision is high. The output results of the 16 groups of test samples are shown in table 2, and it can be seen from table 2 that the fault output codes of the 16 groups of test samples are completely consistent with the preset fault codes, which indicates that the designed neural network can identify faults and has high accuracy.
Table 2: and outputting a result by the test sample.
Figure BDA0003153784770000112
Figure BDA0003153784770000121
Example 2:
the embodiment 2 of the present disclosure provides a power battery fault diagnosis system based on an improved RBF neural network, including:
a data acquisition module configured to: acquiring parameter data of a power battery;
a fault diagnosis module configured to: obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model;
in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm.
The working method of the system is the same as the power battery fault diagnosis method based on the improved RBF neural network provided by the embodiment 1, and the detailed description is omitted here.
Example 3:
the embodiment 3 of the present disclosure provides a computer-readable storage medium, on which a program is stored, which when executed by a processor, implements the steps in the method for diagnosing a fault of a power battery based on an improved RBF neural network according to the embodiment 1 of the present disclosure.
Example 4:
the embodiment 4 of the present disclosure provides an electronic device, which includes a memory, a processor, and a program stored in the memory and executable on the processor, and when the processor executes the program, the steps in the method for diagnosing a fault of a power battery based on an improved RBF neural network according to embodiment 1 of the present disclosure are implemented.
As will be appreciated by one skilled in the art, embodiments of the present disclosure may be provided as a method, system, or computer program product. Accordingly, the present disclosure may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.

Claims (10)

1. A power battery fault diagnosis method based on an improved RBF neural network is characterized in that: the method comprises the following steps:
acquiring parameter data of a power battery;
obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model;
in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm.
2. The improved RBF neural network-based power battery fault diagnosis method of claim 1, wherein:
the parametric data includes at least one or more of battery voltage, battery temperature, SOC, battery internal resistance, and capacity.
3. The improved RBF neural network-based power battery fault diagnosis method of claim 1, wherein:
determining the center of the basis function through a subtractive clustering algorithm, comprising the following processes:
regarding a plurality of input sample data, taking the sample data as a candidate set of a clustering center, and calculating a density index of each data point;
selecting a data point with the highest density index as a first clustering center, and revising the density indexes of other data points by taking the data point as a center;
judging whether the termination criterion of the subtractive clustering algorithm is met;
if the termination criterion of the subtractive clustering algorithm is not met, selecting the point with the highest density index at the moment as a second clustering center, and correcting the density indexes of other data points again by taking the data point as the center;
and repeating the process continuously until the termination criterion of the subtractive clustering algorithm is met, and taking the final clustering center as the center of the basis function.
4. The improved RBF neural network-based power battery fault diagnosis method of claim 3, wherein:
and taking the average distance between the clustering center and the data points adjacent to the clustering center as an expansion coefficient according to the obtained clustering center.
5. The improved RBF neural network-based power battery fault diagnosis method of claim 3, wherein:
the termination criterion of the subtractive clustering algorithm is as follows: and the ratio of the highest density index of all the current data points to the density index of the cluster center is smaller than a preset constant.
6. The improved RBF neural network-based power battery fault diagnosis method of claim 3, wherein:
after the cluster center is reselected, correcting the density indexes of other data points again, wherein the correction formula is as follows:
Figure FDA0003153784760000021
wherein the content of the first and second substances,
Figure FDA0003153784760000022
as a new density index of the cluster center, xiFor the ith data point, the data point,
Figure FDA0003153784760000023
data points, σ, represented by the cluster centersbIs a preset constant.
7. The improved RBF neural network-based power battery fault diagnosis method of claim 3, wherein:
a density indicator for each data point, comprising:
Figure FDA0003153784760000024
where M is the total number of data points, xiIs the ith data point, xjIs the j-th dataPoint, σaIs a preset constant.
8. A power battery fault diagnosis system based on an improved RBF neural network is characterized in that: the method comprises the following steps:
a data acquisition module configured to: acquiring parameter data of a power battery;
a fault diagnosis module configured to: obtaining a fault diagnosis result of the power battery according to the acquired parameter data and a preset RBF neural network model;
in the preset RBF neural network model, a Gaussian function is selected as a basis function, and the center and the expansion coefficient of the basis function are determined through a subtractive clustering algorithm.
9. A computer-readable storage medium, on which a program is stored, wherein the program, when executed by a processor, implements the steps in the improved RBF neural network-based power battery fault diagnosis method according to any one of claims 1-7.
10. An electronic device comprising a memory, a processor and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for diagnosing a fault of a power battery based on an improved RBF neural network as claimed in any one of claims 1 to 7.
CN202110774153.1A 2021-07-08 2021-07-08 Power battery fault diagnosis method and system based on improved RBF neural network Pending CN113325317A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110774153.1A CN113325317A (en) 2021-07-08 2021-07-08 Power battery fault diagnosis method and system based on improved RBF neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110774153.1A CN113325317A (en) 2021-07-08 2021-07-08 Power battery fault diagnosis method and system based on improved RBF neural network

Publications (1)

Publication Number Publication Date
CN113325317A true CN113325317A (en) 2021-08-31

Family

ID=77426036

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110774153.1A Pending CN113325317A (en) 2021-07-08 2021-07-08 Power battery fault diagnosis method and system based on improved RBF neural network

Country Status (1)

Country Link
CN (1) CN113325317A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113884903A (en) * 2021-10-19 2022-01-04 中国计量大学 Battery fault diagnosis method based on multilayer perceptron neural network

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20130074328A (en) * 2011-12-26 2013-07-04 한전케이디엔주식회사 System for predicting a performance of rechargeable battery and drive method of the same
CN103926526A (en) * 2014-05-05 2014-07-16 重庆大学 Analog circuit fault diagnosis method based on improved RBF neural network
CN105823989A (en) * 2016-05-04 2016-08-03 安徽工程大学 Improved battery SOC prediction method for RBF neural network
CN107422266A (en) * 2017-03-15 2017-12-01 中国电力科学研究院 A kind of method for diagnosing faults and device of high capacity cell energy-storage system
CN108445406A (en) * 2018-03-13 2018-08-24 桂林电子科技大学 A kind of power battery health status method of estimation
CN110907732A (en) * 2019-12-04 2020-03-24 江苏方天电力技术有限公司 Phase modulator fault diagnosis method based on PCA-RBF neural network
CN111007401A (en) * 2019-12-16 2020-04-14 国网江苏省电力有限公司电力科学研究院 Electric vehicle battery fault diagnosis method and device based on artificial intelligence
KR20200061704A (en) * 2018-11-26 2020-06-03 현대자동차주식회사 A system for improving fault of a batterty of a vehicle and method thereof
CN111458646A (en) * 2020-05-25 2020-07-28 安徽理工大学 Lithium battery SOC estimation method based on PSO-RBF neural network
CN111652348A (en) * 2020-02-29 2020-09-11 上海电力大学 Power battery pack fault fusion diagnosis method and system based on improved CNN
CN111983468A (en) * 2020-08-24 2020-11-24 哈尔滨理工大学 Safety degree estimation method of lithium power battery based on neural network
DE102019212909A1 (en) * 2019-08-28 2021-03-04 Audi Ag Method for detecting a fault in a battery system as well as battery system and motor vehicle
CN112858917A (en) * 2021-01-15 2021-05-28 哈尔滨工业大学（威海） Battery system multi-fault diagnosis method based on genetic algorithm optimization neural network

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20130074328A (en) * 2011-12-26 2013-07-04 한전케이디엔주식회사 System for predicting a performance of rechargeable battery and drive method of the same
CN103926526A (en) * 2014-05-05 2014-07-16 重庆大学 Analog circuit fault diagnosis method based on improved RBF neural network
CN105823989A (en) * 2016-05-04 2016-08-03 安徽工程大学 Improved battery SOC prediction method for RBF neural network
CN107422266A (en) * 2017-03-15 2017-12-01 中国电力科学研究院 A kind of method for diagnosing faults and device of high capacity cell energy-storage system
CN108445406A (en) * 2018-03-13 2018-08-24 桂林电子科技大学 A kind of power battery health status method of estimation
KR20200061704A (en) * 2018-11-26 2020-06-03 현대자동차주식회사 A system for improving fault of a batterty of a vehicle and method thereof
DE102019212909A1 (en) * 2019-08-28 2021-03-04 Audi Ag Method for detecting a fault in a battery system as well as battery system and motor vehicle
CN110907732A (en) * 2019-12-04 2020-03-24 江苏方天电力技术有限公司 Phase modulator fault diagnosis method based on PCA-RBF neural network
CN111007401A (en) * 2019-12-16 2020-04-14 国网江苏省电力有限公司电力科学研究院 Electric vehicle battery fault diagnosis method and device based on artificial intelligence
CN111652348A (en) * 2020-02-29 2020-09-11 上海电力大学 Power battery pack fault fusion diagnosis method and system based on improved CNN
CN111458646A (en) * 2020-05-25 2020-07-28 安徽理工大学 Lithium battery SOC estimation method based on PSO-RBF neural network
CN111983468A (en) * 2020-08-24 2020-11-24 哈尔滨理工大学 Safety degree estimation method of lithium power battery based on neural network
CN112858917A (en) * 2021-01-15 2021-05-28 哈尔滨工业大学（威海） Battery system multi-fault diagnosis method based on genetic algorithm optimization neural network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘振宇等: "《Battery Fault Diagnosis Scheme Based on Improved RBF Neural Network》", 《2021 IEEE 10TH DATA DRIVEN CONTROL AND LEARNING SYSTEMS CONFERENCE》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113884903A (en) * 2021-10-19 2022-01-04 中国计量大学 Battery fault diagnosis method based on multilayer perceptron neural network
CN113884903B (en) * 2021-10-19 2023-08-18 中国计量大学 Battery fault diagnosis method based on multi-layer perceptron neural network

Similar Documents

Publication Publication Date Title
CN112731159B (en) Method for pre-judging and positioning battery faults of battery compartment of energy storage power station
CN111007401A (en) Electric vehicle battery fault diagnosis method and device based on artificial intelligence
CN112092675A (en) Battery thermal runaway early warning method, system and server
CN113805064B (en) Lithium ion battery pack health state prediction method based on deep learning
CN109001640B (en) Data processing method and device for power battery
CN110568360B (en) Lithium battery aging diagnosis method based on fuzzy logic algorithm
CN113156917A (en) Power grid equipment fault diagnosis method and system based on artificial intelligence
CN111707956B (en) Method for predicting health state and residual life of multi-type lithium ion battery pack management system
CN111965476B (en) Low-voltage diagnosis method based on graph convolution neural network
Zhang et al. Intelligent state of charge estimation of battery pack based on particle swarm optimization algorithm improved radical basis function neural network
CN112632850A (en) Method and system for detecting abnormal battery in lithium battery pack
CN114594378A (en) Lithium battery SOH model training method, prediction method, device and equipment
CN111983474A (en) Lithium ion battery life prediction method and system based on capacity decline model
CN111257753A (en) Battery system fault diagnosis method
CN113406524A (en) Inconsistent fault diagnosis method and system for power battery system
CN113325317A (en) Power battery fault diagnosis method and system based on improved RBF neural network
CN114460481A (en) Energy storage battery thermal runaway early warning method based on Bi-LSTM and attention mechanism
CN115114843A (en) Battery capacity prediction method, related device, equipment and storage medium
CN112327169B (en) Lithium battery residual life prediction method
CN117330987A (en) Method, system, medium and apparatus for time-based battery state of health assessment
Song et al. Capacity estimation method of lithium-ion batteries based on deep convolution neural network
CN112363012A (en) Power grid fault early warning device and method
CN117036103A (en) Electric power system operation situation prediction method based on LSTM (least squares) circulating neural network
Liu et al. Battery fault diagnosis scheme based on improved RBF neural network
CN115128468A (en) Chemical energy storage battery PHM undervoltage fault prediction method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210831