CN117330987B - Method, system, medium and apparatus for time-based battery state of health assessment - Google Patents
Method, system, medium and apparatus for time-based battery state of health assessment Download PDFInfo
- Publication number
- CN117330987B CN117330987B CN202311632004.7A CN202311632004A CN117330987B CN 117330987 B CN117330987 B CN 117330987B CN 202311632004 A CN202311632004 A CN 202311632004A CN 117330987 B CN117330987 B CN 117330987B
- Authority
- CN
- China
- Prior art keywords
- battery
- state
- health
- battery pack
- charge
- 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.)
- Active
Links
- 230000036541 health Effects 0.000 title claims abstract description 135
- 238000000034 method Methods 0.000 title claims abstract description 82
- 238000012216 screening Methods 0.000 claims abstract description 38
- 238000013145 classification model Methods 0.000 claims abstract description 30
- 238000012549 training Methods 0.000 claims abstract description 30
- 238000011156 evaluation Methods 0.000 claims abstract description 20
- 238000007599 discharging Methods 0.000 claims abstract description 10
- 230000008569 process Effects 0.000 claims description 32
- 239000013598 vector Substances 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 18
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 125000004122 cyclic group Chemical group 0.000 claims description 11
- 230000015654 memory Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 7
- 238000012706 support-vector machine Methods 0.000 claims description 4
- KJONHKAYOJNZEC-UHFFFAOYSA-N nitrazepam Chemical compound C12=CC([N+](=O)[O-])=CC=C2NC(=O)CN=C1C1=CC=CC=C1 KJONHKAYOJNZEC-UHFFFAOYSA-N 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 230000006403 short-term memory Effects 0.000 claims description 2
- 230000005611 electricity Effects 0.000 claims 1
- 230000007787 long-term memory Effects 0.000 claims 1
- 230000003862 health status Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000004146 energy storage Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000032683 aging Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000005261 decarburization Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- 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
- G01R31/005—Testing of electric installations on transport means
- G01R31/006—Testing of electric installations on transport means on road vehicles, e.g. automobiles or trucks
- G01R31/007—Testing of electric installations on transport means on road vehicles, e.g. automobiles or trucks using microprocessors or computers
-
- 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
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- 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
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06N3/0499—Feedforward networks
-
- 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
- G06N3/08—Learning methods
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Combustion & Propulsion (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Chemical & Material Sciences (AREA)
- Secondary Cells (AREA)
Abstract
The invention relates to battery state of health assessment, and discloses a method, a system, a medium and equipment for time-based battery state of health assessment, comprising the following steps: setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, and recording the health state of the battery and the charging time of each battery pack passing through each interval step length; calculating the correlation between the health state of the battery and the charging time according to the charging time of each battery pack passing through each interval step, screening and training and adjusting a classification model by using data with high correlation; and acquiring the charging time of each battery pack passing through each interval step length in the battery to be tested, and inputting the trained classification model corresponding to the adjusted screening threshold value to obtain the health state evaluation result of the battery. The invention can realize effective assessment of the health status of the batteries under different working conditions and different types.
Description
Technical Field
The present invention relates to battery state of health assessment, and more particularly to a method, system, medium and apparatus for time-based battery state of health assessment.
Background
State assessment and health management of batteries are one of the hot spots of current energy storage research, and have important significance. The battery has various advantages, such as high energy density, long service life, low self-discharge rate and the like, so the battery has wide application in electric automobiles and grid-connected power storage systems and can become the key of quick energy decarburization transformation in China. In actual use, the performance of the battery may deteriorate due to complex multidimensional aging behavior. To ensure safety, reliability and cost effectiveness during operation, accurate and robust health monitoring of the battery is necessary. Accordingly, more and more research is focused on developing tools and methods for better monitoring battery health.
Methods for estimating and predicting the state of health of a battery can be classified into two types: model-based methods and data-driven based methods. Common model-based methods are electrochemical models, equivalent circuit models, empirical models, and mechanical models, which typically use partial differential equations, different filtering methods, and statistical tools to predict health states in order to model phenomena, states, and parameters related to aging behavior of the energy storage unit. However, the complex interaction process occurring in the battery results in high computational complexity, so that the development and testing of the model-based method have limitations.
In contrast to the model-based approach, the data-driven approach can solve the disadvantage of the model-based approach, which does not require capturing an aging mechanism and detailed internal interactions, and can utilize black box modeling of the input-output functional relationship to predict the health state based on different characteristic health indicators, thereby improving the state evaluation accuracy of the energy storage system. A common data-driven-based method is a method using an artificial neural network, in which a selection of input characteristics and a goal of a model are necessary steps, and an output goal is to display health conditions of an energy storage unit, including health indicators such as a remaining life span, a remaining use period, a ratio of current impedance to initial impedance, a ratio of current capacity to initial capacity, and the like, wherein the capacity-based health indicator is a key indicator for battery energy management and maintenance. However, the data-driven method has a high degree of dependence on the sample battery, and cannot be applied to batteries with different conditions or different chemical properties.
Disclosure of Invention
Therefore, the technical problem to be solved by the invention is to overcome the defects in the prior art, and provide a method, a system, a medium and equipment for estimating the health state of a battery based on time, which can realize effective estimation of the health state of the battery under different working conditions and different types.
In order to solve the above technical problems, the present invention provides a method for estimating a battery state of health based on time, including:
setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, and recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length;
calculating the correlation between the health state of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step length, screening data with high correlation, and dividing the data into a training set and a testing set;
training the classification model by using a training set, and inputting the test set into the trained classification model to obtain the predicted health state of the battery;
calculating the accuracy of the predicted state of health of the battery, and adjusting a screening threshold value when screening data with high correlation according to the accuracy until the accuracy reaches a preset requirement;
and acquiring the charging time of each battery pack passing through each interval step length in the battery to be tested, and inputting the trained classification model corresponding to the adjusted screening threshold value to obtain the health state evaluation result of the battery.
In one embodiment of the present invention, the battery is circularly charged and discharged in the charge state interval, and the state of health of the battery and the charging time of each battery pack passing each interval step in each charging process are recorded, specifically:
recording the charging time of each battery pack passing through each interval step length in each charging process, and combining to obtain the characteristic vector of the charging time of each battery pack in each charging process, wherein the characteristic vector is as follows:
，
wherein,nreference numerals indicating the battery packs in the batteries,irepresents the cycle number of cyclic charge and discharge, X n,i Represent the firstiFeature vector, delta, of charge time of nth battery pack in battery during secondary chargingSThe step size of the interval is indicated,γSrepresents the lower limit of the charge amount in the state of charge interval,Srepresents the upper limit of the charge amount in the state of charge interval,represent the firstiNth battery pack slave in battery during secondary chargingγSTo pass throughjCharging time of individual interval steps.
In one embodiment of the present invention, before calculating the correlation between the state of health of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack in the battery passing through each interval step, the charging time of each battery pack in the battery passing through each interval step is normalized.
In one embodiment of the present invention, when the correlation between the state of health of the battery and the charging time of each battery pack in the battery is calculated according to the charging time of each battery pack passing each interval step, the correlation calculation method used is pearson correlation coefficient method;
the value range of the correlation coefficient obtained by the calculation of the Pelson correlation coefficient method is [ -1,1], when the value of the correlation coefficient is-1, the negative correlation between the health state of the battery and the charging time of each battery pack in the battery is shown; when the value of the correlation coefficient is 1, positive correlation between the health state of the battery and the charging time of each battery pack in the battery is shown, and the correlation is strongest; when the value of the correlation coefficient is 0, it indicates that there is a positive correlation between the state of health of the battery and the charging time of each battery pack in the battery and that the correlation is the weakest.
In one embodiment of the present invention, when the pearson correlation coefficient method is used to calculate the correlation between the state of health of the battery and the charging time of each battery pack in the battery, the calculation formula is:
，
wherein,nreference numerals indicating the battery packs in the batteries,γSrepresents the lower limit of the charge amount in the state of charge interval,Srepresents the upper limit of the charge amount in the state of charge interval,representation ofFrom the slaveγSCharged toSState of health of battery and the first in batterynCorrelation coefficients between charge times of the individual battery packs,iindicates the cycle number of cyclic charge and discharge, +.>Represent the firstiNth battery pack slave in battery during secondary chargingγSCharged toSCharging time of->Indicating the first of the cellsnIndividual battery pack slaveγSCharged toSAverage value of the charging times of all the cyclic charging cycles,/>Represent the firstiIn the secondary charging processnHealth status of individual battery pack->Represent the firstnThe average value of the state of health for all cycle charging cycles of the individual battery packs.
In one embodiment of the invention, the classification model is a feed forward neural network, a recurrent neural network, a convolutional neural network, a long-short term memory network, a support vector machine, a correlation vector machine, gaussian process regression, or ensemble learning.
In one embodiment of the present invention, when the accuracy of the predicted state of health of the battery is calculated, the root mean square percentage error and the average absolute percentage error of the predicted state of health of the battery are used as the evaluation index of the accuracy.
The invention also provides a system for estimating the state of health of the battery based on time, which comprises:
the data acquisition module is used for setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, and recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length;
the data processing module is used for calculating the correlation between the health state of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step length, screening data with high correlation and dividing the data into a training set and a testing set;
the model training module is used for training the classification model by using the training set, and inputting the test set into the classification model after training to obtain the predicted health state of the battery; calculating the accuracy of the predicted state of health of the battery, and adjusting a screening threshold value when screening data with high correlation according to the accuracy until the accuracy reaches a preset requirement;
the evaluation module is used for acquiring the charging time of each battery pack passing through each interval step length in the battery to be tested and inputting the trained classification model corresponding to the adjusted screening threshold value to obtain the health state evaluation result of the battery.
The invention also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the method of battery state of health assessment.
The invention also provides a device for battery state of health assessment comprising a memory, a processor and a computer program stored on the memory and executable on the processor, said processor executing the method for battery state of health assessment as described by the computer program.
Compared with the prior art, the technical scheme of the invention has the following advantages:
1. according to the invention, the battery is circularly charged and discharged to obtain data of charging time and battery health state, the correlation between the charging time and the battery health state is calculated to screen data, a classification model is trained by using the screened data, a screening threshold value is adjusted according to the prediction accuracy of the classification model, parameters of the classification model are re-optimized, and finally the health state of the battery to be detected is predicted by using the classification model after training and optimization; the method can be suitable for batteries with different working conditions and different types, and can be used for effectively predicting the health state of the batteries.
2. The invention extracts the characteristic vector X of the charging time of the characteristic of the health state from the local charging period n,i The reference charging time index can be easily extracted from a part of charging process, and the cycle data is collected and the model is retrained, so that the method can be suitable for batteries with different working conditions, different cycle conditions and different degradation behaviors, and can be used for simultaneously training a single model to predict the health states of all batteries, and the application range is wide.
3. According to the invention, the real-time dynamic evaluation of the health state can be realized by detecting the charging time of the battery to be detected, no additional sensing equipment is needed, the input cost is saved, and the evaluation precision of the health state is greatly improved.
Drawings
In order that the invention may be more readily understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings, in which:
fig. 1 is a flow chart of the method of the present invention.
Fig. 2 is a flow chart of the steps of the method of the present invention.
Fig. 3 is a schematic diagram of correlation analysis of each charge time, battery state of health and state of charge of each battery pack in a battery during a cycle in an embodiment of the present invention.
Fig. 4 is a graph showing the result of battery state of health evaluation in the embodiment of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and specific examples, which are not intended to be limiting, so that those skilled in the art will better understand the invention and practice it.
Referring to fig. 1-2, the invention discloses a method for evaluating the health state of a battery, which comprises the following steps:
s1: setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, recording the state of health (SOH) of the battery and the charging time of each battery pack passing through each interval step length in each charging process, and keeping the experiment environment unchanged each time, namely keeping constant temperature and constant pressure and keeping a uniform charging mode when circularly charging and discharging the battery.
S1-1: when a charge state interval of battery charge and discharge is set, a lower limit and an upper limit of the charge state interval are set, wherein the lower limit is an initial charge state which needs to be chargedγS) The upper limit is the selected state of charge when charging is completeS) The method comprises the steps of carrying out a first treatment on the surface of the The charge time of a battery as a reference during charging may be defined as the time the battery takes from an initial state of charge at which charging is required to a selected state of charge at the completion of charging.
Therefore, the reference charging time can be measured by two parameters, namely the initial state of charge of chargingγS) And the selected state of charge at the completion of chargingS) The initial charge state of the battery with lower charge levelγS) When the charge quantity of the battery reaches the selected state of chargeS) The charging is stopped. The charging time of different references can be obtained by using different initial charge states to be charged and selected charge states when the charging is completed, so that the values of the two parameters must be determined first, and the optimal values of the two parameters depend on two factors: availability of data, correlation between charging time and health status of different references. The former factor means that a range should be selected according to the available charging data, for example: batteries of electric vehicles are often charged from 10% to 80% state of charge, and in order to exclude the effects of anomalous data, a set of reference charge time data should be selected for characterization rather than a single reference charge time data. The latter factor shows that the optimal reference charge time has a reasonable correlation with the deterioration of the health state, so that it is necessary to select the appropriate reference charge time as its health indicator. When the initial state of charge required to be charged is 0, the reference charging time and the health state show strong correlation, but when the electric automobile is used, the state of charge is difficult to start charging from 0%, so that the initial state of charge required to be charged can be selected to be 10%, 20% and the like.
Initial state of charge that requires chargingγS) And complete chargingTime-dependent selected state of charge [ ]S) The value of (2) is required to meetγS<SThe value is selected according to the actual situation, and the difference value between the two parameters should be as large as possible in the process of selecting the two parameters so as to ensure a more accurate health state evaluation result. Taking the above factors into consideration, the smallest is selectedγSCan be 20%, maximumSMay be 80%.
When setting the interval step of battery charge and discharge, the interval step (deltaS) The value of (2) is selected according to the actual situation, and the selection range is 1-1%S-γSIn order to ensure the evaluation accuracy of the state of health of the battery, the step size is selected to be as small as possible, and in this embodiment, the interval step size of each step may be set to be 5%.
S1-2: recording the charging time of each battery pack passing through each interval step length in each charging process, and combining to obtain the characteristic vector of the charging time of each battery pack in each charging process, wherein the characteristic vector is as follows:
，
wherein,nreference numerals indicating the battery packs in the batteries,irepresents the cycle number of cyclic charge and discharge, X n,i Represent the firstiFeature vector, delta, of charge time of nth battery pack in battery during secondary chargingSRepresenting the interval step size i.e. the state of charge change step size,γSrepresents the lower limit of the charge amount in the state of charge interval,Srepresents the upper limit of the charge amount in the state of charge interval,represent the firstiNth battery pack slave in battery during secondary chargingγSTo pass throughjCharging time of individual interval steps.
The battery is subjected to a cyclic charge and discharge experiment, and the experiment process is respectively carried out from the initial charge state needing to be chargedγS) And the selected state of charge at the completion of chargingS) Charge time of recorded referenceWherein->Is the firstiThe reference charging time of the nth battery pack in the battery in the secondary charging process, namely the initial charge state of the battery pack from the need of chargingγS) Selected state of charge by the time charging is completedS) The time required.
The feature vector of the charging time of each battery pack in the battery in each charging process obtained in this embodiment is:
。
s2: and calculating the correlation between the health state of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step length in the battery, screening data with high correlation, and dividing the data into a training set and a test set, namely screening the health state of the battery with high correlation and the charging time of each battery pack passing through each interval step length in the battery as the training set, setting an initial value of a screening threshold value when screening the data with high correlation according to the actual data quantity or an empirical value, and dynamically adjusting the screening threshold value according to accuracy in S4.
S2-1: and normalizing the charging time of each battery pack passing through each interval step length in the battery. The feature vector of the charging time is extracted for each cycle of each battery pack in the dataset, and is used for constructing an input feature matrix of the classification model. In order to facilitate the subsequent calculation of the correlation and training of the classification model, in this embodiment, the feature vectors of the charging time are subjected to matrix normalization, and the value of each column in the feature vectors of the charging time after normalization is within the range of [0,1 ].
S2-2: the pearson correlation coefficient method is used for calculating the correlation between the state of health of the battery and the charging time of each battery pack in the battery, and the calculation formula is as follows:
，
wherein,nreference numerals indicating the battery packs in the batteries,γSrepresents the lower limit of the charge amount in the state of charge interval,Srepresents the upper limit of the charge amount in the state of charge interval,representing slaveγSCharged toSState of health of battery and the first in batterynPearson correlation coefficient between charging times of the individual battery packs,ithe number of cycles of the cyclic charge and discharge is indicated,represent the firstiNth battery pack slave in battery during secondary chargingγSCharged toSIs used for the charging time of the battery,indicating the first of the cellsnIndividual battery pack slaveγSCharged toSAverage value of the charging times of all the cyclic charging cycles,/>Represent the firstiIn the secondary charging processnHealth status of individual battery pack->Represent the firstnThe average value of the state of health for all cycle charging cycles of the individual battery packs.
The value range of the correlation coefficient calculated by the Pelson correlation coefficient method is [ -1,1], when the value of the correlation coefficient is-1, the negative correlation between the health state of the battery and the charging time of each battery pack in the battery is shown; when the value of the correlation coefficient is 1, positive correlation between the health state of the battery and the charging time of each battery pack in the battery is shown, and the correlation is strongest; when the value of the correlation coefficient is 0, positive correlation between the state of health of the battery and the charging time of each battery pack in the battery is shown, and the correlation is the weakest; the closer the limit is, the stronger or weaker the linear correlation between the two is exhibited.
S3: training the classification model by using a training set until the model converges to obtain a trained classification model, and inputting the test set into the trained classification model to obtain the predicted health state of the battery.
The classification model may be a feed forward neural network (feedforward neural network, FNN), a recurrent neural network (recursive neural network, RNN), a convolutional neural network (Convolutional Neural Networks, CNN), a Long Short-Term Memory (LSTM), a support vector machine (Support Vector Machine, SVM), a relevance vector machine (Relevance Vector Machine, RVM), gaussian process regression (Gaussian Process Regression, GPR), or ensemble learning (Ensemble Learning).
The feedforward neural network is simple to implement and has excellent performance in a nonlinear system, so the feedforward neural network is selected as a classification prediction model in this embodiment. For a given data set, a typical feed forward neural network consists of three layers of neurons: an input layer, a hidden layer, and an output layer. The input and output layers are fixed, the existing neurons in these layers are determined according to the number of inputs and outputs, and the number of hidden layers and the underlying neurons are free parameters.
S4: and (3) calculating the accuracy of the predicted state of health of the battery, and adjusting a screening threshold value in the step (S2) when the data with high correlation are screened according to the accuracy until the accuracy reaches a preset requirement.
Taking the root mean square percentage error and the average absolute percentage error of the predicted state of health of the battery as an evaluation index of accuracy:
root mean square percentage errorRMSPEThe calculation method of (1) is as follows:
，
wherein,Nindicating the total number of battery packs in the battery,representing predicted in-battery itemnHealth status of individual battery pack->An average value representing the predicted state of health of all battery packs in the battery;
mean absolute percentage errorMAPEThe calculation method of (1) is as follows:。
if the accuracy corresponding to the adjusted next screening threshold is higher than that corresponding to the previous screening threshold, reducing the number of data with high correlation screened by the screening threshold; if the accuracy corresponding to the next screening threshold is lower than that corresponding to the previous screening threshold, the number of data with high correlation screened by the screening threshold is enlarged; and (5) ending the adjustment until the accuracy is not changed or the maximum number of times of the preset screening threshold value when the data with high correlation is screened is reached.
S5: after the screening threshold is adjusted, the charging time of each battery pack passing through each interval step in the battery to be tested is obtained, and the trained classification model corresponding to the adjusted screening threshold is input to obtain the battery health state evaluation result.
The invention also discloses a system for evaluating the health state of the battery, which comprises a data acquisition module, a data processing module, a model training module and an evaluation module.
The data acquisition module is used for setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, and recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length. The data processing module is used for calculating the correlation between the health state of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step length, screening data with high correlation and dividing the data into a training set and a testing set. The model training module is used for training the classification model by using the training set, and inputting the test set into the classification model after training to obtain the predicted health state of the battery; and calculating the accuracy of the predicted state of health of the battery, and adjusting a screening threshold value when screening the data with high correlation according to the accuracy until the accuracy reaches a preset requirement. The evaluation module is used for acquiring the charging time of each battery pack passing through each interval step length in the battery to be tested and inputting the trained classification model corresponding to the adjusted screening threshold value to obtain the health state evaluation result of the battery.
The invention also discloses a computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method of battery state of health assessment.
The invention also discloses a device for evaluating the health state of the battery, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the method for evaluating the health state of the battery when executing the computer program.
To further illustrate the beneficial effects of the present invention, in this embodiment, the charging time, the battery state of health and the state of charge of each battery pack in the battery during the cycle are analyzed in a correlation manner, and the results are shown in fig. 3. In fig. 3, the abscissa indicates the charging time, the ordinate indicates the State of Charge (SOC) of the battery, and different broken lines indicate the relationship between the charging time and the battery State of Charge under different battery states of health (SOH). As can be seen from fig. 3, there is an association among the charging time, the battery state of health and the state of charge, so the practice of the present invention is feasible.
Meanwhile, in this embodiment, a simulation experiment is performed on the battery by using the method of the present invention, and the actual health status is compared with the estimated health status, and the result is shown in fig. 4. The abscissa in fig. 4 is the actual health state, the ordinate is the health state predicted using the present invention, the straight line is the ideal state, that is, the estimated health state is identical to the actual health state, and the discrete point in fig. 4 is the relationship between the predicted health state and the actual health state value obtained in the simulation experiment. As can be seen from fig. 4, the battery state of health predicted by the present invention is substantially consistent with the actual state of health, and the prediction accuracy of the battery state of health is high.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application 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 application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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 is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations and modifications of the present invention will be apparent to those of ordinary skill in the art in light of the foregoing description. It is not necessary here nor is it exhaustive of all embodiments. And obvious variations or modifications thereof are contemplated as falling within the scope of the present invention.
Claims (8)
1. A method of battery state of health assessment, comprising:
setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, and recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length;
calculating the correlation between the health state of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step length, screening data with high correlation, and dividing the data into a training set and a testing set;
training the classification model by using a training set, and inputting the test set into the trained classification model to obtain the predicted health state of the battery;
calculating the accuracy of the predicted state of health of the battery, and adjusting a screening threshold value when screening data with high correlation according to the accuracy until the accuracy reaches a preset requirement;
acquiring charging time of each battery pack passing through each interval step length in the battery to be tested, and inputting a trained classification model corresponding to the adjusted screening threshold value to obtain a battery health state evaluation result;
the method comprises the steps of circularly charging and discharging the battery in a charge state interval, recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length in the battery, and specifically comprises the following steps:
recording the charging time of each battery pack passing through each interval step length in each charging process, and combining to obtain the characteristic vector of the charging time of each battery pack in each charging process, wherein the characteristic vector is as follows:
wherein n represents the number of battery packs in the battery, i represents the cycle number of cyclic charge and discharge, X n,i Feature vector representing charging time of nth battery pack in battery in ith charging process, Δs representing the section step length, γs representing lower limit of charge amount of charge state section, S representing upper limit of charge amount of charge state section, RCT n，i (γs, γs+jΔs) represents the charging time from γs to the passage of j interval steps of the nth battery pack in the battery during the ith charging process;
when the correlation between the state of health of the battery and the charging time of each battery pack in the battery is calculated according to the charging time of each battery pack passing through each interval step length, the used correlation calculation method is a pearson correlation coefficient method; when the pearson correlation coefficient method is used for calculating the correlation between the state of health of the battery and the charging time of each battery pack in the battery, the calculation formula is as follows:
where n represents the label of the battery pack in the battery, γs represents the lower limit of the charge amount in the state of charge section, S represents the upper limit of the charge amount in the state of charge section, ρ n (γS, S) represents a correlation coefficient between the state of health of the battery when charged from γS to S and the charging time of the nth battery pack in the battery, i represents the cycle number of cyclic charge and discharge, RCT n，i (γS, S) represents the charging time from γS to S of the nth battery pack in the battery during the ith charging,indicating that the nth battery pack in the battery is charged from gamma SAverage of charge time for all cycle charge cycles of electricity to S, SOH n，i Indicating the state of health of the nth battery pack during the ith charge, +.>The average value of the state of health of all the cycle charging cycles of the nth battery pack is represented.
2. The method of battery state of health assessment of claim 1, wherein: before calculating the correlation between the state of health of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step, carrying out normalization processing on the charging time of each battery pack passing through each interval step.
3. The method of battery state of health assessment of claim 1, wherein: the value range of the correlation coefficient obtained by the calculation of the Pelson correlation coefficient method is [ -1,1], when the value of the correlation coefficient is-1, the negative correlation between the health state of the battery and the charging time of each battery pack in the battery is shown; when the value of the correlation coefficient is 1, positive correlation between the health state of the battery and the charging time of each battery pack in the battery is shown, and the correlation is strongest; when the value of the correlation coefficient is 0, it indicates that there is a positive correlation between the state of health of the battery and the charging time of each battery pack in the battery and that the correlation is the weakest.
4. The method of battery state of health assessment of claim 1, wherein: the classification model is a feedforward neural network, a recurrent neural network, a convolution neural network, a long-term and short-term memory network, a support vector machine, a correlation vector machine, gaussian process regression or ensemble learning.
5. The method for battery state of health assessment according to any one of claims 1-4, wherein: and when the accuracy of the predicted state of health of the battery is calculated, taking the root mean square percentage error and the average absolute percentage error of the predicted state of health of the battery as an accuracy evaluation index.
6. A system for battery state of health assessment, comprising:
the data acquisition module is used for setting a charge state interval and interval step length of battery charge and discharge, circularly charging and discharging the battery in the charge state interval, and recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length;
the data processing module is used for calculating the correlation between the health state of the battery and the charging time of each battery pack in the battery according to the charging time of each battery pack passing through each interval step length, screening data with high correlation and dividing the data into a training set and a testing set;
the model training module is used for training the classification model by using the training set, and inputting the test set into the classification model after training to obtain the predicted health state of the battery; calculating the accuracy of the predicted state of health of the battery, and adjusting a screening threshold value when screening data with high correlation according to the accuracy until the accuracy reaches a preset requirement;
the evaluation module is used for acquiring the charging time of each battery pack passing through each interval step length in the battery to be tested and inputting the trained classification model corresponding to the adjusted screening threshold value to obtain the health state evaluation result of the battery;
the method comprises the steps of circularly charging and discharging the battery in a charge state interval, recording the health state of the battery in each charging process and the charging time of each battery pack passing through each interval step length in the battery, and specifically comprises the following steps:
recording the charging time of each battery pack passing through each interval step length in each charging process, and combining to obtain the characteristic vector of the charging time of each battery pack in each charging process, wherein the characteristic vector is as follows:
wherein n represents the number of battery packs in the battery, i represents the cycle number of cyclic charge and discharge, X n,i Feature vector representing charging time of nth battery pack in battery in ith charging process, Δs representing the section step length, γs representing lower limit of charge amount of charge state section, S representing upper limit of charge amount of charge state section, RCT n，i (γs, γs+jΔs) represents the charging time from γs to the passage of j interval steps of the nth battery pack in the battery during the ith charging process;
when the correlation between the state of health of the battery and the charging time of each battery pack in the battery is calculated according to the charging time of each battery pack passing through each interval step length, the used correlation calculation method is a pearson correlation coefficient method; when the pearson correlation coefficient method is used for calculating the correlation between the state of health of the battery and the charging time of each battery pack in the battery, the calculation formula is as follows:
where n represents the label of the battery pack in the battery, γs represents the lower limit of the charge amount in the state of charge section, S represents the upper limit of the charge amount in the state of charge section, ρ n (γS, S) represents a correlation coefficient between the state of health of the battery when charged from γS to S and the charging time of the nth battery pack in the battery, i represents the cycle number of cyclic charge and discharge, RCT n，i (γS, S) represents the charging time from γS to S of the nth battery pack in the battery during the ith charging,average value of charge time representing all cycle charge cycles of charging from γs to S of nth battery pack in battery, SOH n，i Indicating the state of health of the nth battery pack during the ith charge, +.>The average value of the state of health of all the cycle charging cycles of the nth battery pack is represented.
7. A computer-readable storage medium having stored thereon a computer program, characterized by: the computer program, when executed by a processor, implements the method of battery state of health assessment of any of claims 1-5.
8. An apparatus for battery state of health assessment, characterized by: comprising a memory, a processor and a computer program stored on the memory and executable on the processor, said processor implementing the method of battery state of health assessment according to any of claims 1-5 when said computer program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311632004.7A CN117330987B (en) | 2023-12-01 | 2023-12-01 | Method, system, medium and apparatus for time-based battery state of health assessment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311632004.7A CN117330987B (en) | 2023-12-01 | 2023-12-01 | Method, system, medium and apparatus for time-based battery state of health assessment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117330987A CN117330987A (en) | 2024-01-02 |
| CN117330987B true CN117330987B (en) | 2024-02-20 |
Family
ID=89279783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311632004.7A Active CN117330987B (en) | 2023-12-01 | 2023-12-01 | Method, system, medium and apparatus for time-based battery state of health assessment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117330987B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117805656A (en) * | 2024-01-08 | 2024-04-02 | 杭州科工电子科技股份有限公司 | Method and system for measuring health of battery pack |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108226805A (en) * | 2018-01-18 | 2018-06-29 | 武汉理工大学 | A kind of cell health state On-line Estimation method based on the charging stage |
| JP2018107992A (en) * | 2016-12-26 | 2018-07-05 | エンコアード テクノロジーズ インク | Power usage information collection system and method |
| CN113158345A (en) * | 2021-04-29 | 2021-07-23 | 浙江吉利控股集团有限公司 | New energy vehicle power battery capacity prediction method and system |
| CN113917337A (en) * | 2021-10-13 | 2022-01-11 | 国网福建省电力有限公司 | Battery health state estimation method based on charging data and LSTM neural network |
| CN114578237A (en) * | 2022-02-28 | 2022-06-03 | 西安交通大学 | Method, system and equipment for rapidly estimating health state of lithium ion battery based on constant-current charging time |
| CN114936505A (en) * | 2022-03-18 | 2022-08-23 | 福州大学 | Method for rapidly forecasting multi-point water depth of urban rainwater well |
| CN115248390A (en) * | 2021-11-22 | 2022-10-28 | 昆明理工大学 | Lithium ion battery SOH estimation method based on random short-term charging data |
| CN115598557A (en) * | 2022-08-26 | 2023-01-13 | 广东工业大学（Cn） | Lithium battery SOH estimation method based on constant voltage charging current |
| CN115982141A (en) * | 2022-12-03 | 2023-04-18 | 哈尔滨工程大学 | Characteristic optimization method for time series data prediction |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9111212B2 (en) * | 2011-08-19 | 2015-08-18 | Hartford Steam Boiler Inspection And Insurance Company | Dynamic outlier bias reduction system and method |
| US10557840B2 (en) * | 2011-08-19 | 2020-02-11 | Hartford Steam Boiler Inspection And Insurance Company | System and method for performing industrial processes across facilities |
| US9069725B2 (en) * | 2011-08-19 | 2015-06-30 | Hartford Steam Boiler Inspection & Insurance Company | Dynamic outlier bias reduction system and method |
| CN115032541B (en) * | 2022-04-13 | 2023-12-05 | 北京理工大学 | Lithium ion battery capacity degradation prediction method and device |
-
2023
- 2023-12-01 CN CN202311632004.7A patent/CN117330987B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018107992A (en) * | 2016-12-26 | 2018-07-05 | エンコアード テクノロジーズ インク | Power usage information collection system and method |
| CN108226805A (en) * | 2018-01-18 | 2018-06-29 | 武汉理工大学 | A kind of cell health state On-line Estimation method based on the charging stage |
| CN113158345A (en) * | 2021-04-29 | 2021-07-23 | 浙江吉利控股集团有限公司 | New energy vehicle power battery capacity prediction method and system |
| CN113917337A (en) * | 2021-10-13 | 2022-01-11 | 国网福建省电力有限公司 | Battery health state estimation method based on charging data and LSTM neural network |
| CN115248390A (en) * | 2021-11-22 | 2022-10-28 | 昆明理工大学 | Lithium ion battery SOH estimation method based on random short-term charging data |
| CN114578237A (en) * | 2022-02-28 | 2022-06-03 | 西安交通大学 | Method, system and equipment for rapidly estimating health state of lithium ion battery based on constant-current charging time |
| CN114936505A (en) * | 2022-03-18 | 2022-08-23 | 福州大学 | Method for rapidly forecasting multi-point water depth of urban rainwater well |
| CN115598557A (en) * | 2022-08-26 | 2023-01-13 | 广东工业大学（Cn） | Lithium battery SOH estimation method based on constant voltage charging current |
| CN115982141A (en) * | 2022-12-03 | 2023-04-18 | 哈尔滨工程大学 | Characteristic optimization method for time series data prediction |
Non-Patent Citations (3)
| Title |
|---|
| Guangze LI ; et..State of Health Prediction for Battery Based on Ensemble Learning.2021 International Conference on Electronics,Circuits and Information Engineering..2021,全文. * |
| 基于激光诱导击穿光谱技术寻优定量分析土壤中Mn元素;沙文;李江涛;鲁翠萍;;中国激光;20200510(05);全文 * |
| 陈彦余,等.基于EMD-ARMA的锂离子电池剩余寿命预测.电力学报.2021,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117330987A (en) | 2024-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112731159B (en) | Method for pre-judging and positioning battery faults of battery compartment of energy storage power station | |
| KR102354112B1 (en) | Apparatus and method for estimating status of battery based on artificial intelligence | |
| JP5313250B2 (en) | Battery long-term characteristic prediction system and method | |
| JP7497432B2 (en) | Battery performance prediction | |
| CN117330987B (en) | Method, system, medium and apparatus for time-based battery state of health assessment | |
| CN111832221B (en) | Lithium battery life prediction method based on feature screening | |
| JP2010538246A (en) | Battery long-term characteristic prediction system and method | |
| Qin et al. | Prognostics of remaining useful life for lithium-ion batteries based on a feature vector selection and relevance vector machine approach | |
| CN113189495B (en) | Battery health state prediction method and device and electronic equipment | |
| CN114818831B (en) | Bidirectional lithium ion battery fault detection method and system based on multi-source perception | |
| US20220170995A1 (en) | Method and Apparatus for Predicting a State of Health of a Device Battery in a Battery-operated Device | |
| CN111983474A (en) | Lithium ion battery life prediction method and system based on capacity decline model | |
| CN114280479A (en) | Electrochemical impedance spectrum-based rapid sorting method for retired batteries | |
| CN113030761A (en) | Method and system for evaluating health state of battery of super-large-scale energy storage power station | |
| CN113447828A (en) | Lithium battery temperature estimation method and system based on Bayesian neural network | |
| Sohn et al. | Two-stage deep learning for online prediction of knee-point in Li-ion battery capacity degradation | |
| CN112881916A (en) | Method and system for predicting health state and remaining usable life of lithium battery | |
| Yao et al. | Fault identification of lithium-ion battery pack for electric vehicle based on ga optimized ELM neural network | |
| Hatherall et al. | Remaining discharge energy estimation for lithium-ion batteries using pattern recognition and power prediction | |
| Sohn et al. | CNN-based online diagnosis of knee-point in Li-ion battery capacity fade curve | |
| Zhang et al. | Remaining useful life prediction of lithium-ion batteries based on TCN-DCN fusion model combined with IRRS filtering | |
| CN115267586A (en) | Lithium battery SOH evaluation method | |
| CN114879043A (en) | Lithium ion battery lithium analysis diagnosis method, device, equipment and medium | |
| Li et al. | State of health estimation and prediction of electric vehicle power battery based on operational vehicle data | |
| Saleem et al. | Investigation of Deep Learning Based Techniques for Prognostic and Health Management of Lithium-Ion Battery |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |