CN114881318A - Lithium battery health state prediction method and system based on generation countermeasure network - Google Patents
Lithium battery health state prediction method and system based on generation countermeasure network Download PDFInfo
- Publication number
- CN114881318A CN114881318A CN202210461560.1A CN202210461560A CN114881318A CN 114881318 A CN114881318 A CN 114881318A CN 202210461560 A CN202210461560 A CN 202210461560A CN 114881318 A CN114881318 A CN 114881318A
- Authority
- CN
- China
- Prior art keywords
- neural network
- lithium battery
- prediction
- encoder
- model
- 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.)
- Granted
Links
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 title claims abstract description 55
- 229910052744 lithium Inorganic materials 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000036541 health Effects 0.000 title claims abstract description 39
- 238000003062 neural network model Methods 0.000 claims abstract description 69
- 238000013528 artificial neural network Methods 0.000 claims abstract description 28
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000009826 distribution Methods 0.000 claims description 27
- 238000005070 sampling Methods 0.000 claims description 13
- 239000000126 substance Substances 0.000 claims description 8
- 230000003862 health status Effects 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 12
- 238000003860 storage Methods 0.000 description 14
- 238000004590 computer program Methods 0.000 description 9
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 229910001416 lithium ion Inorganic materials 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000006731 degradation reaction Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003487 electrochemical reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005184 irreversible process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003446 memory effect Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007784 solid electrolyte Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a lithium battery health state prediction method and system based on a generated countermeasure network. Integrating the optimal prediction result as a lithium battery health state comprehensive prediction model by calculating the mutual support degree of the basic neural network prediction result and the variational self-encoder generation result; and calculating the health state of the lithium battery by using a lithium battery health state comprehensive prediction model according to various parameters in the actual charging process of the lithium battery. Because the invention can generate more basic neural network model prediction results by using a small number of basic neural network models and the variational self-encoder, the diversity of the basic learner in the integrated learning process is ensured, the model training process is accelerated, and the prediction precision close to that of the traditional integrated learning method can be obtained.
Description
Technical Field
The invention belongs to the field of lithium ion battery health state evaluation, and particularly relates to a lithium battery health state prediction method and system based on a generation countermeasure network.
Background
Lithium ion batteries have been the core energy supply components of many electric devices due to their superior properties such as light weight, high energy density, no memory effect, and low self-discharge rate. However, the charge and discharge process of the lithium ion battery can form a solid electrolyte intermediate phase, which seriously affects the electrochemical reaction inside the lithium ion battery, and the irreversible process can lead to the continuous attenuation of the capacity of the lithium ion battery. When the lithium battery is continuously subjected to charge and discharge cycles, chemical components of the lithium battery tend to age, the safety of the whole system is influenced by the performance degradation of the battery, and if the State of Health (SOH) of the lithium battery cannot be accurately evaluated and measures such as maintenance or replacement are taken in time, catastrophic consequences can be caused, so that serious economic loss and even casualties are caused. Therefore, the lithium battery should be retired when the lithium battery is degraded to a certain degree, so as to ensure the safety and reliability of the lithium battery power supply system. The indexes for evaluating the performance degradation of the lithium battery comprise battery capacity, output power, internal resistance and the like, the health state of the lithium battery can be estimated by predicting the change trend of the indexes, and the battery is maintained in time, so that powerful support is provided for the reliability of system operation, and the method has important significance for the safety of an industrial system.
In recent years, with the rapid increase of computer computing power, Neural Network (Neural Network) technology has been rapidly developed, and has attracted considerable attention in the fields of image processing, natural language processing, and the like. Therefore, the neural network technology has great practical significance and application prospect in the aspect of predicting the health state of the lithium ion battery. However, the degradation process of the lithium battery has high uncertainty, and due to the structural characteristics of a single neural network, all data characteristics cannot be captured, so that high precision is difficult to achieve. Therefore, it is necessary to adopt various heterogeneous neural networks for prediction, and integrate the prediction results of various network models by an ensemble learning technique to obtain a better prediction result.
In order to enable a lithium ion battery to safely and efficiently operate on electric equipment, a multi-network-fused lithium ion battery health state prediction method is urgently needed to solve the problem of low accuracy of a single network prediction result.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a lithium battery health state prediction method and system based on a generation countermeasure network, and solves the problem of low accuracy of the conventional single network prediction result.
The invention is realized by the following technical scheme:
a lithium battery health state prediction method based on generation of a countermeasure network comprises the following steps:
step 1, constructing and training a variational self-encoder model, respectively taking prediction results of a plurality of pre-trained basic neural network models as real samples input by the variational self-encoder model, wherein the prediction results output by the trained variational self-encoder model have certain similarity with the corresponding real sample distribution;
step 2, determining the support degrees of any two prediction results in the prediction results of the plurality of basic neural network models and the prediction results generated by the variational self-encoder, calculating and determining the initial weight of each basic neural network model and the variational self-encoder according to the support degrees, determining the weighted prediction results of each basic neural network model and the variational self-encoder model according to the initial weight, and obtaining the integrated prediction results of the plurality of neural network models according to each weighted prediction result;
step 3, determining a weight adjustment coefficient of each basic neural network model according to the integrated prediction result and the support degree of each basic neural network model, and adjusting the initial weight of each basic neural network model according to the weight adjustment coefficient to obtain the adjusted weight of each basic neural network model;
and 4, repeatedly executing the step 2 and the step 3 according to the adjusted weight of each basic neural network model until the preset iteration times are reached to obtain a lithium battery health state comprehensive prediction network model, and predicting the health state of the lithium battery according to the lithium battery health state comprehensive prediction network model.
Preferably, the plurality of basic neural network models includes a convolutional neural network model, a residual network model, and a gated recursive network model.
Preferably, the variational self-encoder model comprises an encoder and a decoder, wherein the encoder comprises a mean neural network f 1 And variance neural network f 2 The decoder is a multilayer perceptron;
the encoder is used for determining the normal distribution of each real sample;
the decoder is used for reconstructing the normal distribution by adopting samples, outputting the prediction result of the variational self-encoder model and ensuring that the prediction result has certain similarity with the real sample distribution.
Preferably, the training method of the variational auto-encoder model is as follows:
s1, inputting the prediction result of the basic neural network model into an encoder as a real sample, and calculating the mean value and the variance of the real sample by using the mean neural network and the variance neural network;
s1, determining normal distribution of each real sample according to the mean value and the variance of the real samples;
s2, sampling a sampling sample from the normal distribution, and reconstructing the sampling sample by a decoder;
s3, minimizing the error between the sampling sample and the reconstructed sampling sample;
s4, adjusting the network parameters of the encoder and the decoder according to the error, and iteratively executing the step S1 to the step S4 until the loss is lower than a certain threshold value, thereby finishing the training of the variational self-encoder.
Preferably, the mean neural network f 1 And variance neural network f 2 The expression of (a) is as follows:
μ k ＝f 1 (X k )
wherein, X k Representing the kth true sample; mu.s k Means representing a normal distribution of real samples;
representing the variance of the true sample normal distribution.
Preferably, the error is expressed as follows:
wherein the content of the first and second substances,
represents the loss after fusion, d represents the dimension of the hidden variable Z in the variational autocoder, mu (i) Represents the ith component of a normally distributed mean vector,
represents the ith component of a normally distributed variance vector.
Preferably, the expression of the support degree in step 2 is as follows:
wherein the content of the first and second substances,
representing the predicted results of two different underlying neural network models.
Preferably, the initial weight expression of the basic neural network model in step 2 is as follows:
preferably, the expression of the weight adjustment coefficient in step 3 is as follows:
wherein the content of the first and second substances,
and E (A) represents the prediction results of two different basic neural network models, and E (A) is an integrated prediction result.
A lithium battery health state prediction system based on a generation countermeasure network executes the steps of the lithium battery health state prediction method based on the generation countermeasure network when the system runs.
Compared with the prior art, the invention has the following beneficial technical effects:
the invention provides a lithium battery health state prediction method based on a generation countermeasure network, which comprises a pre-trained basic neural network model, a variational self-encoder and an integrated learning method based on a support degree. Firstly, the prediction result of the basic neural network model is input into the variational self-encoder, so that the generation result of the variational self-encoder and the prediction result distribution of the basic neural network model have certain similarity. Integrating the optimal prediction result as a lithium battery health state comprehensive prediction model by calculating the mutual support degree of the basic neural network prediction result and the variational self-encoder generation result; and calculating the health state of the lithium battery by using the comprehensive lithium battery health state prediction model according to various parameters in the actual charging process of the lithium battery. Because the invention can generate more basic neural network model prediction results by using a small number of basic neural network models and the variational self-encoder, the diversity of the basic learning device in the integrated learning process is ensured, compared with the traditional prediction method adopting the integrated learning, the invention can reduce the model training cost, accelerate the model training process and obtain the prediction precision close to the traditional integrated learning method.
Drawings
Fig. 1 is a flow chart of a lithium battery health status prediction method based on generation of a countermeasure network according to the present invention.
Detailed Description
The present invention will now be described in further detail with reference to the attached drawings, which are illustrative, but not limiting, of the present invention.
Referring to fig. 1, the method for predicting the health status of a lithium battery based on a generated countermeasure network provided by the invention comprises the following steps:
step 1, according to data collected in a lithium battery cyclic charge-discharge process and a plurality of preset basic neural network models;
the collected data are voltage, current, temperature and state of charge data of the lithium battery of the model in the cyclic charge and discharge process.
The preset multiple basic Neural Network models include a Convolutional Neural Network (CNN) for extracting local features, a Residual Network (ResNet) for extracting timing features, and a Gated recursive Network (GRU) for extracting timing features.
Step 2, constructing a variational self-encoder and training, wherein the variational self-encoder comprises an encoder and a decoder, the encoder comprises a mean value calculation network and a variance calculation network, the decoder is of a multilayer perceptron, a prediction result of a basic neural network model is used as the input of the encoder, and finally the prediction result output by the decoder and the prediction result input by the encoder have certain similarity by adjusting parameters of the encoder and the decoder, namely the variational self-encoder can generate a result similar to the distribution of the basic neural network model, and the specific method comprises the following steps:
s21, constructing a variational self-encoder model, which comprises an encoder and a decoder, wherein the encoder comprises a mean value neural network f 1 And variance neural network f 2 Calculating the normal distribution of each input sample according to the output result of the encoder; the decoder is a multilayer perceptron, and finally a loss function of the variational self-encoder is set, and parameters of the encoder and the decoder are adjusted according to the loss function.
S22, training the variational self-encoder model, inputting the prediction result of the basic neural network model as a real sample to the encoder, and inputting the mean neural network f 1 And variance neural network f 2 Calculating the mean value mu of the real sample k Sum variance σ k 2 。
The mean neural network f 1 And variance neural network f 2 The expression of (a) is as follows:
μ k ＝f 1 (X k )
wherein, X k Representing the kth input real sample; mu.s k Representing a mean neural network f 1 Calculating the mean value of normal distribution of real samples;
representation by variance neural network f 2 And calculating the variance of the normal distribution of the real sample.
S23, according to the real sample X k Mean value of (a) k Sum variance
The normal distribution of a real sample can be assumed to be:
sampling a sample Z from the normal distribution k The sampling sample Z k After the sample is input into a decoder g, a reconstructed sample can be obtained
I.e. the prediction result of the variational self-encoder model, the expression of the decoder network g is as follows:
s24 minimization of sample X k And reconstructed sample
The reconstructed sample output by the decoder has a certain similarity with the real sample distribution of the input of the encoder, and the error is the fused reconstruction loss and the KL divergence and can be measured by the following formula:
wherein
Represents the loss after fusion; d represents the dimension of a hidden variable Z in the variational self-encoder; mu.s (i) Represents the ith component of the normal distribution mean vector;
represents the ith component of a normally distributed variance vector.
S25, adjusting the network parameters of the encoder and the decoder according to the error, and iteratively executing the step S22 to the step S24 until the loss is lower than a certain threshold value, thereby completing the construction of the variational self-encoder.
In this embodiment, the prediction results of the multiple basic neural network models are input to the encoder to generate normal distribution of multiple real samples, after sampling and decoding by the decoder, the reconstruction loss and KL divergence between the output of the decoder and the input of the encoder are calculated, and iteration is performed until the loss is lower than a certain threshold, thereby completing the construction of the variational self-encoder.
And 3, determining the support degree between any two prediction results in the prediction results of the plurality of basic neural network models and the prediction results generated by the variational self-encoder to obtain a plurality of support degrees, calculating the initial weight of each prediction result according to the plurality of support degrees to obtain the initial weight of each basic neural network model and the variational self-encoder, multiplying the prediction results of each basic neural network model and the variational self-encoder by the corresponding initial weight to obtain the weighted prediction results of each basic neural network model and the variational self-encoder, and adding all the weighted prediction results to obtain an integrated prediction result.
In this embodiment, three basic neural network models are preset, three prediction results are input into the variational self-encoder, three prediction results are obtained, six prediction results are obtained in total, the support degree between every two prediction results is calculated, and the initial weight of each prediction result is calculated according to the obtained support degrees.
Defining the prediction results of the plurality of basic neural network models and the prediction results of the variational self-encoder as follows:
the weights corresponding to the plurality of neural network prediction results are expressed as:
W＝{ω i |i＝1,2,...,n}
wherein, ω is i Need to satisfy
Representing the predicted results of two different underlying neural network models, then
And
the support between is defined as follows:
wherein the content of the first and second substances,
normalized euclidean distance:
prediction of any underlying neural network model
The support of (2) is defined as follows:
according to the definition of the support degree,
the corresponding initial weight calculation formula is as follows:
after the initial weight of each basic neural network model is obtained, the integrated prediction results E (A) of a plurality of basic neural network models are weighted and calculated:
E(A)＝AW T
and 4, determining a weight adjustment coefficient of each basic neural network model according to the integrated prediction result and the support degree of each basic neural network model, and carrying out self-adaptive dynamic adjustment on the initial weight of each basic neural network model according to the weight adjustment coefficient to obtain the adjusted weight of each basic neural network model. So as to adapt to the characteristics of different data and further improve the integration performance.
ω i Predicting results for a neural network model
Wherein i is 1, 2. And E (A) is the prediction result of the integrated neural network model calculated by using the arithmetic weighted average, and the weight adjustment coefficient is defined as follows:
the weights are updated using the following formula:
wherein, ω is i (k) Represents the coefficient, omega, of the ith network model in the last iteration process i (k+1) Expressing the adjusted coefficient of the ith network model, obtaining new weight, and obtaining the new weight according to E (A) ═ AW T And performing weighted calculation on the integrated prediction result.
And 5, repeatedly executing the step 3 and the step 4 according to the adjusted weight of each basic neural network model until the preset iteration times are reached, and obtaining the comprehensive lithium battery health state prediction network model.
And calculating the health state of the lithium battery by using a lithium battery health state comprehensive prediction model according to various parameters in the actual charging process of the lithium battery.
A lithium battery health state prediction system based on generation of a countermeasure network comprises,
the model construction module is used for constructing a variational self-encoder model, respectively taking the prediction results of a plurality of pre-trained basic neural network models as real samples input by the variational self-encoder model, and the prediction results output by the trained variational self-encoder model have certain similarity with the corresponding real sample distribution;
the integrated prediction module is used for determining the support degrees of any two prediction results in the prediction results of the plurality of basic neural network models and the prediction results generated by the variational self-encoder, calculating and determining the initial weight of each basic neural network model and the corresponding variational self-encoder according to the support degrees, determining the weighted prediction results of each basic neural network model and the variational self-encoder model according to the initial weight, and obtaining the integrated prediction results of the plurality of neural network models according to each weighted prediction result;
the integrated prediction module is used for determining a weight adjustment coefficient of each basic neural network model according to an integrated prediction result and the support degree of each basic neural network model, and adjusting the initial weight of each basic neural network model according to the weight adjustment coefficient to obtain the adjusted weight of each basic neural network model;
and the comprehensive prediction module is used for iteratively updating the adjusted weight until the preset iteration times are reached, so as to obtain the lithium battery health state comprehensive prediction network model.
In yet another embodiment of the present invention, a computer device is provided that includes a processor and a memory for storing a computer program comprising program instructions, the processor for executing the program instructions stored by the computer storage medium. The Processor may be a Central Processing Unit (CPU), or may be other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable gate array (FPGA) or other Programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., which is a computing core and a control core of the terminal, and is specifically adapted to load and execute one or more instructions in a computer storage medium to implement a corresponding method flow or a corresponding function; the processor of the embodiment of the invention can be used for the operation of the lithium battery health state prediction method based on the generation of the countermeasure network.
In yet another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a Memory device in a computer device and is used for storing programs and data. It is understood that the computer readable storage medium herein can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also, one or more instructions, which may be one or more computer programs (including program code), are stored in the memory space and are adapted to be loaded and executed by the processor. It should be noted that the computer-readable storage medium may be a high-speed RAM memory, or may be a non-volatile memory (non-volatile memory), such as at least one disk memory. One or more instructions stored in the computer-readable storage medium may be loaded and executed by the processor to implement the corresponding steps of the method for predicting the health status of the lithium battery based on the generation of the countermeasure network in the above embodiments.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.
Claims (10)
1. A lithium battery health state prediction method based on generation of a countermeasure network is characterized by comprising the following steps:
step 1, constructing and training a variational self-encoder model, respectively taking prediction results of a plurality of pre-trained basic neural network models as real samples input by the variational self-encoder model, wherein the prediction results output by the trained variational self-encoder model have certain similarity with the corresponding real sample distribution;
step 2, determining the support degrees of any two prediction results in the prediction results of the plurality of basic neural network models and the prediction results generated by the variational self-encoder, calculating and determining the initial weight of each basic neural network model and the variational self-encoder according to the support degrees, determining the weighted prediction results of each basic neural network model and the variational self-encoder model according to the initial weight, and obtaining the integrated prediction results of the plurality of neural network models according to each weighted prediction result;
step 3, determining a weight adjustment coefficient of each basic neural network model according to the integrated prediction result and the support degree of each basic neural network model, and adjusting the initial weight of each basic neural network model according to the weight adjustment coefficient to obtain the adjusted weight of each basic neural network model;
and 4, repeatedly executing the step 2 and the step 3 according to the adjusted weight of each basic neural network model until the preset iteration times are reached to obtain a lithium battery health state comprehensive prediction network model, and predicting the health state of the lithium battery according to the lithium battery health state comprehensive prediction network model.
2. The lithium battery state of health prediction method based on generation of countermeasure network of claim 1, wherein the plurality of basic neural network models includes a convolutional neural network model, a residual network model and a gated recursive network model.
3. The lithium battery health state prediction method based on generation countermeasure network as claimed in claim 1, wherein the variational self-encoder model comprises two parts of encoder and decoder, the encoder comprises mean neural network f 1 And variance neural network f 2 The decoder is a multilayer perceptron;
the encoder is used for determining the normal distribution of each real sample;
the decoder is used for reconstructing the normal distribution by adopting samples, outputting the prediction result of the variational self-encoder model and ensuring that the prediction result has certain similarity with the real sample distribution.
4. The lithium battery state of health prediction method based on generation countermeasure network as claimed in claim 3, characterized in that the training method of the variational self-encoder model is as follows:
s1, inputting the prediction result of the basic neural network model into an encoder as a real sample, and calculating the mean value and the variance of the real sample by using the mean neural network and the variance neural network;
s1, determining the normal distribution of each real sample according to the mean value and the variance of the real samples;
s2, sampling a sampling sample from the normal distribution, and reconstructing the sampling sample by a decoder;
s3, minimizing the error between the sampling sample and the reconstructed sampling sample;
s4, adjusting the network parameters of the encoder and the decoder according to the error, and iteratively executing the step S1 to the step S4 until the loss is lower than a certain threshold value, thereby finishing the training of the variational self-encoder.
5. The lithium battery health state prediction method based on generation countermeasure network as claimed in claim 4, wherein the mean neural network f 1 And variance neural network f 2 The expression of (a) is as follows:
μ k ＝f 1 (X k )
wherein, X k Representing the kth true sample; mu.s k Means representing a normal distribution of real samples;
represents the variance of the true sample normal distribution.
6. The lithium battery health status prediction method based on generation countermeasure network as claimed in claim 4, wherein the expression of the error is as follows:
wherein the content of the first and second substances,
represents the loss after fusion, d represents the dimension of the hidden variable Z in the variational autocoder, mu (i) Represents the ith component of a normally distributed mean vector,
represents the ith component of a normally distributed variance vector.
7. The lithium battery health state prediction method based on generation of countermeasure network as claimed in claim 1, wherein the expression of the support degree in step 2 is as follows:
wherein the content of the first and second substances,
representing the predicted results of two different underlying neural network models.
8. The lithium battery health state prediction method based on generation of the countermeasure network as claimed in claim 7, wherein the initial weight expression of the basic neural network model in step 2 is as follows:
9. the lithium battery health status prediction method based on generation of countermeasure network as claimed in claim 1, wherein the expression of the weight adjustment coefficient in step 3 is as follows:
wherein the content of the first and second substances,
and E (A) represents the prediction results of two different basic neural network models, and E (A) is an integrated prediction result.
10. A lithium battery health status prediction system based on a generation countermeasure network, characterized in that the system is operated to execute the steps of the lithium battery health status prediction method based on the generation countermeasure network according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210461560.1A CN114881318B (en) | 2022-04-28 | 2022-04-28 | Lithium battery health state prediction method and system based on generation countermeasure network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210461560.1A CN114881318B (en) | 2022-04-28 | 2022-04-28 | Lithium battery health state prediction method and system based on generation countermeasure network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114881318A true CN114881318A (en) | 2022-08-09 |
| CN114881318B CN114881318B (en) | 2024-07-05 |
Family
ID=82670740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210461560.1A Active CN114881318B (en) | 2022-04-28 | 2022-04-28 | Lithium battery health state prediction method and system based on generation countermeasure network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114881318B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298935A (en) * | 2023-05-19 | 2023-06-23 | 河南科技学院 | Lithium ion battery health state estimation method based on countermeasure encoder network |
| CN116466243A (en) * | 2023-06-16 | 2023-07-21 | 国网安徽省电力有限公司电力科学研究院 | Method and device for evaluating health state of lithium battery based on generation countermeasure network |
| CN116500460A (en) * | 2023-06-29 | 2023-07-28 | 北京云控安创信息技术有限公司 | Cloud computing-based battery health state diagnosis and prediction system for Internet of things |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102018002714A1 (en) * | 2017-04-18 | 2018-10-18 | Gabriele Trinkel | Memristor effect system network and process with functional material |
| CN108705928A (en) * | 2017-04-11 | 2018-10-26 | 邦迪克斯商用车***有限责任公司 | Use the mixed powered vehicle heat management of dynamic heater |
| CN111584816A (en) * | 2019-02-19 | 2020-08-25 | 奥迪股份公司 | Protective device, battery, motor vehicle and method for switching off a battery cell |
| CN112834945A (en) * | 2020-12-31 | 2021-05-25 | 东软睿驰汽车技术（沈阳）有限公司 | Evaluation model establishing method, battery health state evaluation method and related product |
| CN112904370A (en) * | 2019-11-15 | 2021-06-04 | 辉达公司 | Multi-view deep neural network for lidar sensing |
| CN112926505A (en) * | 2021-03-24 | 2021-06-08 | 重庆大学 | Rotating machine health index construction method based on DTC-VAE neural network |
| CN113033678A (en) * | 2021-03-31 | 2021-06-25 | 温州大学 | Lithium battery pack fault diagnosis method based on adaptive countermeasure network |
| CN113408138A (en) * | 2021-06-29 | 2021-09-17 | 广东工业大学 | Lithium battery SOH estimation method and system based on secondary fusion |
| CN113810543A (en) * | 2021-08-09 | 2021-12-17 | 荣耀终端有限公司 | Signal processing method and related device for adjusting interference and interference rejection in terminal equipment |
| CN113920010A (en) * | 2020-07-10 | 2022-01-11 | 华为技术有限公司 | Super-resolution implementation method and device for image frame |
| CN114004336A (en) * | 2021-10-29 | 2022-02-01 | 上海科技大学 | Three-dimensional ray reconstruction method based on enhanced variational self-encoder |
-
2022
- 2022-04-28 CN CN202210461560.1A patent/CN114881318B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108705928A (en) * | 2017-04-11 | 2018-10-26 | 邦迪克斯商用车***有限责任公司 | Use the mixed powered vehicle heat management of dynamic heater |
| DE102018002714A1 (en) * | 2017-04-18 | 2018-10-18 | Gabriele Trinkel | Memristor effect system network and process with functional material |
| CN111584816A (en) * | 2019-02-19 | 2020-08-25 | 奥迪股份公司 | Protective device, battery, motor vehicle and method for switching off a battery cell |
| CN112904370A (en) * | 2019-11-15 | 2021-06-04 | 辉达公司 | Multi-view deep neural network for lidar sensing |
| CN113920010A (en) * | 2020-07-10 | 2022-01-11 | 华为技术有限公司 | Super-resolution implementation method and device for image frame |
| CN112834945A (en) * | 2020-12-31 | 2021-05-25 | 东软睿驰汽车技术（沈阳）有限公司 | Evaluation model establishing method, battery health state evaluation method and related product |
| CN112926505A (en) * | 2021-03-24 | 2021-06-08 | 重庆大学 | Rotating machine health index construction method based on DTC-VAE neural network |
| CN113033678A (en) * | 2021-03-31 | 2021-06-25 | 温州大学 | Lithium battery pack fault diagnosis method based on adaptive countermeasure network |
| CN113408138A (en) * | 2021-06-29 | 2021-09-17 | 广东工业大学 | Lithium battery SOH estimation method and system based on secondary fusion |
| CN113810543A (en) * | 2021-08-09 | 2021-12-17 | 荣耀终端有限公司 | Signal processing method and related device for adjusting interference and interference rejection in terminal equipment |
| CN114004336A (en) * | 2021-10-29 | 2022-02-01 | 上海科技大学 | Three-dimensional ray reconstruction method based on enhanced variational self-encoder |
Non-Patent Citations (3)
| Title |
|---|
| ZHANG HENG SHAN 等: "support-based neural network ensemble method for predicting the soH of lithium-ion battery", PROCEEDINGS OF THE 2022 5TH INTERNATIONAL CONFERENCE ONARTIFICIAL INTELLIGENCE AND PATTERN RECOGNITION, 30 September 2022 (2022-09-30), pages 1 - 7 * |
| 任罡等: "基于VMD和Bat-KELM的仿真变电站蓄电池剩余寿命预测", 电源学报, 25 April 2022 (2022-04-25), pages 13 * |
| 王廷银;林明贵;陈达;吴允平;: "基于北斗RDSS的核辐射监测应急通讯方法", 计算机***应用, vol. 18, no. 12, 15 December 2019 (2019-12-15), pages 248 - 252 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298935A (en) * | 2023-05-19 | 2023-06-23 | 河南科技学院 | Lithium ion battery health state estimation method based on countermeasure encoder network |
| CN116298935B (en) * | 2023-05-19 | 2023-09-19 | 河南科技学院 | Lithium ion battery health state estimation method based on countermeasure encoder network |
| CN116466243A (en) * | 2023-06-16 | 2023-07-21 | 国网安徽省电力有限公司电力科学研究院 | Method and device for evaluating health state of lithium battery based on generation countermeasure network |
| CN116500460A (en) * | 2023-06-29 | 2023-07-28 | 北京云控安创信息技术有限公司 | Cloud computing-based battery health state diagnosis and prediction system for Internet of things |
| CN116500460B (en) * | 2023-06-29 | 2023-08-22 | 北京云控安创信息技术有限公司 | Cloud computing-based battery health state diagnosis and prediction system for Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114881318B (en) | 2024-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114881318B (en) | Lithium battery health state prediction method and system based on generation countermeasure network | |
| Yang et al. | Remaining useful life prediction of lithium-ion batteries based on a mixture of ensemble empirical mode decomposition and GWO-SVR model | |
| CN112036084B (en) | Similar product life migration screening method and system | |
| CN108647839A (en) | Voltage-stablizer water level prediction method based on cost-sensitive LSTM Recognition with Recurrent Neural Network | |
| CN111832825B (en) | Wind power prediction method and system integrating long-term memory network and extreme learning machine | |
| Vidal et al. | Estimating battery state of charge using recurrent and non-recurrent neural networks | |
| Zheng et al. | State of health estimation for lithium battery random charging process based on CNN-GRU method | |
| CN111784061B (en) | Training method, device and equipment for power grid engineering cost prediction model | |
| CN115616414A (en) | Method and device for predicting remaining life long-time sequence of energy storage lithium battery | |
| CN114757340A (en) | Lithium battery health state prediction method and system based on neural network integration | |
| Li et al. | A hybrid framework for predicting the remaining useful life of battery using Gaussian process regression | |
| Biggio et al. | Dynaformer: A deep learning model for ageing-aware battery discharge prediction | |
| CN114660497A (en) | Lithium ion battery service life prediction method aiming at capacity regeneration phenomenon | |
| Guo et al. | Prognostics of lithium-ion batteries health state based on adaptive mode decomposition and long short-term memory neural network | |
| Hong et al. | State-of-health estimation of lithium-ion batteries using a novel dual-stage attention mechanism based recurrent neural network | |
| CN117406100A (en) | Lithium ion battery remaining life prediction method and system | |
| CN113689039A (en) | Lithium ion battery RUL time sequence prediction method | |
| CN111831955A (en) | Lithium ion battery residual life prediction method and system | |
| CN115528750B (en) | Power grid safety and stability oriented data model hybrid drive unit combination method | |
| CN113884936B (en) | ISSA coupling DELM-based lithium ion battery health state prediction method | |
| Fu et al. | Lithium-ion battery SOH prediction based on VMD-PE and improved DBO optimized temporal convolutional network model | |
| Adducul et al. | Edge-based Battery Remaining Useful Life Estimation Using Deep Learning | |
| CN111626472A (en) | Scene trend judgment index computing system and method based on deep hybrid cloud model | |
| Lee et al. | Attention Mechanism for Lithium-Ion Battery Lifespan Prediction: Temporal and Cyclic Attention | |
| Liao et al. | Lithium batteries State of Health estimation based on optimized SAE-Transformer network |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhang Hengshan Inventor after: Kang Zijun Inventor after: Han Zhaoyan Inventor after: Xu Jiaxuan Inventor after: Wang Yun Inventor after: Mi Jishi Inventor after: Li Haoru Inventor after: Zhou Yun Inventor after: Li Baoxuan Inventor after: Han Haiyang Inventor after: Zhu Zijie Inventor before: Xu Jiaxuan Inventor before: Han Zhaoyan Inventor before: Zhang Hengshan Inventor before: Wang Yun Inventor before: Mi Jishi Inventor before: Li Haoru Inventor before: Zhou Yun Inventor before: Li Baoxuan Inventor before: Han Haiyang Inventor before: Zhu Zijie Inventor before: Kang Zijun |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant |