CN114910793A - SOH estimation method for series battery pack of energy storage power station - Google Patents
SOH estimation method for series battery pack of energy storage power station Download PDFInfo
- Publication number
- CN114910793A CN114910793A CN202210435672.XA CN202210435672A CN114910793A CN 114910793 A CN114910793 A CN 114910793A CN 202210435672 A CN202210435672 A CN 202210435672A CN 114910793 A CN114910793 A CN 114910793A
- Authority
- CN
- China
- Prior art keywords
- battery pack
- discharge
- voltage
- battery
- terminal voltage
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—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
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
The invention discloses a SOH estimation method of a series battery pack of an energy storage power station, which comprises the following steps: performing an off-line discharge test on the battery pack to construct an original characteristic data set; according to the original data, further calculating to extract features, and combining the features with the original feature data set to form a new feature data set; respectively adopting two methods to carry out feature screening on the new feature data set, and reserving an intersection as a feature data subset; preprocessing the characteristic data subset; dividing the characteristic data subset; constructing a Gaussian regression model; training a Gaussian regression model; the SOH of the energy storage plant battery is estimated. According to the method, the SOH of the battery pack can be estimated in the actual operation condition of the energy storage power station by extracting the effective information of all the battery monomers in the battery pack and taking the intersection of the two characteristic screening methods as characteristic data without depending on the full-charging ideal test condition, so that the model calculation amount is reduced, and the accuracy and the real-time performance of the SOH estimation method are improved.
Description
Technical Field
The invention relates to the technical field of battery energy storage, in particular to a SOH estimation method for a series battery pack of an energy storage power station.
Background
In order to meet the requirements of safe operation and reliable supply of electric energy of a novel power system, a large-scale battery energy storage system needs to be introduced into a power grid to assist in realizing frequency modulation and peak shaving. In order to ensure that a battery energy storage power station can operate safely and well, state of health (SOH) estimation needs to be performed on a battery pack of the energy storage power station. In energy storage power stations, the battery pack is usually composed of a plurality of cells connected in series. Because the energy storage power station usually needs to participate in peak clipping and valley filling and active frequency support, the battery pack is frequently switched between charge and discharge modes, and the battery pack can rarely operate in a full-charge and full-discharge working condition. The conventional data-driven battery SOH estimation method is generally specific to a single battery, is usually tested under an ideal working condition based on full charge and discharge, and is difficult to be directly applied to SOH estimation of a battery pack of an energy storage power station. In addition, in the existing estimation method for the SOH of the battery pack, the extracted features generally only include the total current and the total voltage of the battery pack, the features of the single batteries are not extracted, and the effective information of the single batteries in the battery pack cannot be fully utilized, so that the SOH estimation accuracy of the battery pack is low.
Disclosure of Invention
The invention provides an energy storage power station battery pack SOH estimation method for overcoming the problems that the conventional energy storage power station battery pack SOH estimation method needs to depend on an ideal working condition of full charge and discharge for testing and has insufficient estimation precision.
The primary objective of the present invention is to solve the above technical problems, and the technical solution of the present invention is as follows:
s1: the method comprises the steps of carrying out an off-line discharge test on a battery pack in an energy storage power station, enabling the battery pack to pause discharge for multiple times in the process of discharging the battery pack from a charge cut-off voltage to a discharge cut-off voltage, collecting voltage data of the whole process that terminal voltage of all battery monomers in the battery pack at a sampling moment before the battery monomers pause discharge, the terminal voltage at the discharge pause moment rises instantly, and the terminal voltage after the discharge pause recovers to a steady state value, calculating an SOH value of the battery pack according to total discharge capacity, and combining the SOH value with the SOH value before the discharge pauseThe discharge current and terminal voltage of the battery pack are constructed to form an original characteristic data set D ori ；
S2: the raw feature data set D acquired according to step S1 ori Calculating the range and standard deviation of the terminal voltage value at the sampling moment before the suspension of the discharge of all the battery monomers, calculating the standard deviation of the instantaneous rise value of the terminal voltage at the suspension of the discharge of all the battery monomers, calculating the overall process voltage recovery rate of the terminal voltage of all the battery monomers after the suspension of the discharge to the steady state value, calculating the range and standard deviation of the steady state terminal voltage value of all the battery monomers, and comparing the range and standard deviation with the original characteristic data set D in the step S1 ori Merging to form a new feature data set D new ；
S3: applying Pearson' S correlation coefficient method and recursive feature elimination method to the new feature data set D in step S2 new Carrying out feature screening, reserving the common features obtained by screening the two methods, and forming a feature data subset D sub ；
S4: subset D of feature data sub Dividing the training set and the test set into a training set and a test set, and carrying out normalization processing;
s5: constructing a Gaussian regression model;
s6: using the training set data for model training, outputting the model if the condition is met, otherwise, repeating S4-S6 until the model meets the condition;
s7: during the operation of the energy storage power station, when a certain battery pack exits the discharging operation, the current and the terminal voltage of the battery pack are collected, the voltage data of the whole process that the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharging is stopped, the instantaneous rise value of the terminal voltage at the discharging stopping moment and the terminal voltage after the discharging is stopped are recovered to the steady state value are collected, and the characteristic data subset D of the step S3 is extracted from the collected data by adopting the method of the step S2 sub And inputting the corresponding characteristic quantity into a trained Gaussian regression model, and estimating the SOH of the battery pack.
In this embodiment, the original feature data set D of step S1 ori The method comprises the following specific construction steps:
s1-1: the allowable current interval [0, I ] of the battery pack max ]Performing equal difference division for N times, and obtaining a current data set comprising N current values as follows:
I＝[I 1 ,I 2 ,...,I N ]
s1-2: recording the number of single batteries contained in the battery pack as K, and setting the charging and discharging cut-off voltage range [ U ] of the battery pack pack-dis ，U pack-cha ]Performing M-order arithmetic division as voltage values of the suspended discharge point, obtaining a voltage data set including M voltage values as follows:
U＝[U 1 ,U 2 ,...,U M ]
s1-3: from the current data set I ═ I 1 ,I 2 ,...,I N ]The method comprises the steps of selecting the ith current value to perform constant current discharge on a battery pack, pausing discharge for M times in the period, collecting discharge current and terminal voltage of the battery pack when the discharge is paused each time, collecting the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharge is paused, the instantaneous rise value of the terminal voltage at the discharge pausing moment, and the voltage data of the whole process that the terminal voltage is recovered to the steady state value after the discharge pausing, calculating SOH of the battery pack according to the total discharge when the discharge cutoff voltage is reached, constructing original data based on the obtained data, and regarding the mth pause discharge original data of the ith current, wherein the form is as follows:
s1-4: step S1-3 is executed in a loop until the voltage data set U ═ U during the discharge of the battery pack at the ith current is acquired 1 ，U 2 ，...，U M ]The original data corresponding to all M voltage values in the data set constitute the following original data subset D ori，i ：
S1-5: from the current data set I ═ I 1 ，I 2 ，...，I N ]Sequentially selecting all current values, and circularly executing the steps S1-3 to S1-4 to obtain the NxM current value of the battery packThe raw data of the sub-pause discharge constitute the following raw data set D ori ：
In this embodiment, the new feature data set D of step S2 new The method comprises the following specific steps:
s2-1: extracting the range of the terminal voltage value at the sampling moment before the discharge of all the single batteries is suspended as the characteristic according to the m-th suspension discharge original data of the battery pack at the ith current, and recording the range as the characteristic
S2-2: calculating the standard deviation of the terminal voltage value at the sampling moment before the discharge of all the single batteries is suspended as the characteristic according to the m-th suspended discharge original data of the battery pack at the ith current, and recording the standard deviation as the characteristic
S2-3: calculating the standard deviation of the instantaneous terminal voltage rise at the discharge suspension moment of all the single batteries according to the m-th discharge suspension original data of the battery pack at the ith current, and recording the standard deviation as
S2-4: calculating the whole process voltage recovery rate of the voltage recovery to a steady state value after the discharge of all the single batteries is suspended according to the m times of the original discharge suspension data of the battery pack at the ith current as the characteristic, and recording the whole process voltage recovery rate as the characteristic
S2-5: calculating the standard deviation of the steady-state terminal voltage values of all the battery cells as the characteristic according to the m-th suspended discharge original data of the battery pack at the ith current, and recording the standard deviation as the characteristic
S2-6: the features calculated in steps S2-1 to S2-5 are combined with the original data D of step S1-3 ori，i，m And combining to obtain a new characteristic data form of the battery pack in the m-th suspended discharge at the ith current as follows:
s2-7: steps S2-1 to S2-6 are cyclically executed until all the new characteristic data of the battery pack discharged M times at the ith current value are acquired, and the following new characteristic data subset D is formed new，i ：
S2-8: from the current data set I ═ I 1 ，I 2 ，...，I N ]Sequentially selecting all current values, circularly executing the steps S2-1 to S2-7, and obtaining new characteristic data of the battery pack in N multiplied by M times of discharge suspension to form a new characteristic data set D new ：
In this solution, the feature screening method described in step S3 includes the following specific steps:
s3-1: the new feature data set D obtained in step S2 new The method comprises the steps of sequentially carrying out Pearson correlation coefficient test on each feature and the battery pack SOH, reserving the feature with the absolute value of the correlation number being more than 0.8, and removing the rest features to obtain a feature subset D sub-1 The specific calculation formula of the correlation coefficient is as follows:
wherein, the value range of r is [ -1, 1], when r is more than 0, the variables are linearly and positively correlated; when r < 0, the variables are linearly inversely correlated; when r is 0, there is no linear relationship between the variables.
S3-2: the new feature data set D obtained in step S2 new Inputting the initial feature subset into a random forest model, calculating the importance of each feature, calculating the estimation accuracy of the initial feature subset by using a cross-validation method, and removing the feature with the minimum importance from the initial feature subset to obtain a new feature subset; the operation is circulated to iterate until all the features are traversed, and according to the sequence of feature screening, the first 6 features are selected as high-quality features to form a feature subset D sub-2 ；
S3-3: feature subset D determined in steps S3-1 and S3-2 sub-1 And D sub-2 The common feature is preserved and constitutes the optimal feature subset D sub 。
In this embodiment, when a certain battery pack exits the discharging operation in step S7, the SOH of the battery pack is estimated, and the specific steps are as follows:
s7-1: importing the qualified Gaussian regression model trained in the step S6 and the kernel function parameters thereof into a battery pack management system of the energy storage power station;
s7-2: the battery pack management system of the energy storage power station collects and monitors the voltage and the current of all battery packs and all battery monomers in the battery packs in real time, when a certain battery pack exits from discharging operation, the current and the terminal voltage of the battery pack are collected, the voltage data of the whole process that the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharge is stopped, the instantaneous rise value of the terminal voltage at the discharge stopping moment and the terminal voltage after the discharge is stopped are collected, and the characteristic data subset D of the step S3 is extracted from the collected data by adopting the method of the step S2 sub A corresponding characteristic quantity;
s7-3: for the feature data subset D obtained in step S7-2 sub The result of the normalization process is input to the gaussian regression model introduced in step S7-1, and is estimated, thereby obtaining the SOH of the battery pack.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the battery SOH estimation method provided by the invention does not need to rely on a full-charge ideal test working condition, can estimate the SOH of the battery pack in actual operation working conditions such as peak clipping, valley filling, active frequency support and the like, and is more suitable for the actual working conditions of the energy storage power station; the SOH estimation precision of the battery pack is improved by extracting effective information of all battery monomers in the battery pack; by taking the intersection of the two feature screening methods as feature data, the calculated amount is reduced, and the real-time performance of the SOH estimation method is improved.
Drawings
Fig. 1 is a flow chart of an SOH estimation method for a series battery of an energy storage power station according to the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
In a specific embodiment, as shown in fig. 1, a method for estimating SOH of a series battery of energy storage power stations includes the following steps:
s1: the method comprises the steps of carrying out an off-line discharge test on a battery pack in an energy storage power station, enabling the battery pack to pause discharge for multiple times in the process of discharging the battery pack from a charge cut-off voltage to a discharge cut-off voltage, collecting voltage data of the whole process that all battery monomers in the battery pack are recovered to a steady state value after the voltage data of the battery cells are collected at a sampling moment before the battery cells pause discharge, the instantaneous rise value of the terminal voltage at the discharge pause moment and the voltage data of the battery cells are recovered to the steady state value after the battery cells pause discharge, calculating the SOH value of the battery pack according to the total discharge capacity, and constructing an original characteristic data set D by combining the discharge current and the terminal voltage of the battery pack before the battery cells pause discharge ori ；
S2: according to step S1Original feature data set D of the set ori Calculating the range and standard deviation of the terminal voltage value at the sampling moment before the suspension of the discharge of all the battery monomers, calculating the standard deviation of the instantaneous rise value of the terminal voltage at the suspension of the discharge of all the battery monomers, calculating the overall process voltage recovery rate of the terminal voltage of all the battery monomers after the suspension of the discharge to the steady state value, calculating the range and standard deviation of the steady state terminal voltage value of all the battery monomers, and comparing the range and standard deviation with the original characteristic data set D in the step S1 ori Merging to form a new characteristic data set D new ；
S3: applying Pearson' S correlation coefficient method and recursive feature elimination method to the new feature data set D in step S2 new Carrying out feature screening, reserving the common features obtained by screening the two methods, and forming a feature data subset D sub ；
S4: subset D of feature data sub Dividing the training set and the test set into a training set and a test set, and carrying out normalization processing;
s5: constructing a Gaussian regression model;
s6: using the training set data for model training, outputting the model if the condition is met, otherwise, repeating S4-S6 until the model meets the condition;
s7: during the operation of the energy storage power station, when a certain battery pack exits the discharging operation, the current and the terminal voltage of the battery pack are collected, the voltage data of the whole process that the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharging is stopped, the instantaneous rise value of the terminal voltage at the discharging stopping moment and the terminal voltage after the discharging is stopped are recovered to the steady state value are collected, and the characteristic data subset D of the step S3 is extracted from the collected data by adopting the method of the step S2 sub And inputting the corresponding characteristic quantity into a trained Gaussian regression model, and estimating the SOH of the battery pack.
In this embodiment, the original feature data set D of step S1 ori The method comprises the following specific construction steps:
s1-1: the allowable current interval [0, I ] of the battery pack max ]Performing equal difference division for N times, and obtaining a current data set comprising N current values as follows:
I＝[I 1 ,I 2 ,...,I N ]
s1-2: recording the number of single batteries contained in the battery pack as K, and setting the charging and discharging cut-off voltage range [ U ] of the battery pack pack-dis ，U pack-cha ]Performing M-order arithmetic division as voltage values of the suspended discharge point, obtaining a voltage data set including M voltage values as follows:
U＝[U 1 ,U 2 ,...,U M ]
s1-3: from the current data set I ═ I 1 ,I 2 ,...,I N ]The method comprises the steps of selecting the ith current value to perform constant current discharge on a battery pack, pausing discharge for M times in the period, collecting discharge current and terminal voltage of the battery pack when the discharge is paused each time, collecting the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharge is paused, the instantaneous rise value of the terminal voltage at the discharge pausing moment, and the voltage data of the whole process that the terminal voltage is recovered to the steady state value after the discharge pausing, calculating SOH of the battery pack according to the total discharge when the discharge cutoff voltage is reached, constructing original data based on the obtained data, and regarding the mth pause discharge original data of the ith current, wherein the form is as follows:
s1-4: step S1-3 is executed in a loop until the voltage data set U ═ U during the discharge of the battery pack at the ith current is acquired 1 ,U 2 ,...,U M ]The original data corresponding to all M voltage values in the data set constitute the following original data subset D ori,i ：
S1-5: from the current data set I ═ I 1 ,I 2 ,...,I N ]Sequentially selecting all current values, circularly executing the steps S1-3 to S1-4, and obtaining the original data of the battery pack in N multiplied by M times of discharge pause to form the following original data set D ori ：
In this embodiment, the new feature data set D of step S2 new The method comprises the following specific steps:
s2-1: extracting the range of the terminal voltage value at the sampling moment before the discharge of all the single batteries is suspended as the characteristic according to the m-th suspension discharge original data of the battery pack at the ith current, and recording the range as the characteristic
S2-2: calculating the standard deviation of the terminal voltage value at the sampling moment before the discharge of all the single batteries is suspended as the characteristic according to the m-th suspended discharge original data of the battery pack at the ith current, and recording the standard deviation as the characteristic
S2-3: calculating the standard deviation of the instantaneous terminal voltage rise at the discharge suspension moment of all the single batteries according to the m-th discharge suspension original data of the battery pack at the ith current, and recording the standard deviation as
S2-4: calculating the overall process voltage recovery rate of the terminal voltage of all the battery monomers after the discharge is suspended to a steady value according to the m-th suspended discharge original data of the battery pack at the ith current as a characteristic, and recording the overall process voltage recovery rate as
S2-5: calculating the standard deviation of the steady-state terminal voltage values of all the battery cells as the characteristic according to the m-th suspended discharge original data of the battery pack at the ith current, and recording the standard deviation as the characteristic
S2-6: the features calculated in steps S2-1 to S2-5 are combined with the original data D of step S1-3 ori,i,m And combining to obtain a new characteristic data form of the battery pack in the m-th suspended discharge at the ith current as follows:
s2-7: steps S2-1 to S2-6 are cyclically executed until all new characteristic data of the battery pack discharged M times at the ith current value are acquired, constituting a new characteristic data subset D new,i ：
S2-8: from the current data set I ═ I 1 ,I 2 ,...,I N ]Sequentially selecting all current values, circularly executing the steps S2-1 to S2-7, and obtaining new characteristic data of the battery pack for N multiplied by M times of suspended discharge to form a new characteristic data set D new ：
In this scheme, the feature screening method in step S3 includes the following specific steps:
s3-1: the new feature data set D obtained in step S2 new The method comprises the steps of sequentially carrying out Pearson correlation coefficient inspection on each characteristic and the SOH of the battery pack, reserving the characteristic with the absolute value of the correlation number being more than 0.8, and removing the rest characteristics to obtain a characteristic subset D sub-1 The specific calculation formula of the correlation coefficient is as follows:
wherein, the value range of r is [ -1, 1], when r is more than 0, the variables are linearly and positively correlated; when r < 0, the variables are linearly inversely correlated; when r is 0, there is no linear relationship between the variables.
S3-2: new obtained in step S2Feature data set D new Inputting the initial feature subset into a random forest model, calculating the importance of each feature, calculating the estimation accuracy of the initial feature subset by using a cross-validation method, and removing the feature with the minimum importance from the initial feature subset to obtain a new feature subset; the operation is circulated to iterate until all the features are traversed, and according to the sequence of feature screening, the first 6 features are selected as high-quality features to form a feature subset D sub-2 ；
S3-3: feature subset D determined in steps S3-1 and S3-2 sub-1 And D sub-2 The common feature is preserved and constitutes the optimal feature subset D sub 。
In this embodiment, when a certain battery pack exits the discharging operation in step S7, the SOH of the battery pack is estimated, and the specific steps are as follows:
s7-1: importing the qualified Gaussian regression model trained in the step S6 and the kernel function parameters thereof into a battery pack management system of the energy storage power station;
s7-2: the battery pack management system of the energy storage power station collects and monitors the voltage and current of all battery packs and all battery monomers in the battery packs in real time, when a certain battery pack exits from discharging operation, the current and terminal voltage of the battery pack are collected, the voltage data of the whole process that the terminal voltage of all battery monomers in the battery pack at the sampling moment before the suspension of discharging, the instantaneous rise value of the terminal voltage at the suspension discharging moment and the terminal voltage of all battery monomers in the battery pack recover to the steady state value after the suspension of discharging are collected, and the method of the step S2 is adopted to extract the characteristic data subset D of the step S3 from the collected data sub A corresponding characteristic quantity;
s7-3: for the feature data subset D obtained in step S7-2 sub After normalization processing, the result is input to the gaussian regression model introduced in step S7-1 and estimated, so as to obtain the SOH of the battery pack.
Claims (5)
1. A SOH estimation method of a series battery pack of an energy storage power station is characterized by comprising the following specific steps:
s1: the battery pack is subjected to off-line discharge test in the energy storage power station, and the battery pack is chargedIn the process of discharging from the electric cut-off voltage to the discharge cut-off voltage, the battery pack is stopped discharging for multiple times, the voltage data of the whole process that the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharge is stopped, the instantaneous rise value of the terminal voltage at the discharge stopping moment and the terminal voltage after the discharge is stopped are recovered to the steady state value are collected in the process, the SOH value of the battery pack is calculated according to the total discharge amount, and an original characteristic data set D is constructed by combining the discharge current and the terminal voltage of the battery pack before the discharge is stopped ori ；
S2: the raw feature data set D acquired according to step S1 ori Calculating the range and standard deviation of the terminal voltage value at the sampling moment before the suspension of the discharge of all the battery monomers, calculating the standard deviation of the instantaneous rise value of the terminal voltage at the suspension of the discharge of all the battery monomers, calculating the overall process voltage recovery rate of the terminal voltage of all the battery monomers after the suspension of the discharge to the steady state value, calculating the range and standard deviation of the steady state terminal voltage value of all the battery monomers, and comparing the range and standard deviation with the original characteristic data set D in the step S1 ori Merging to form a new feature data set D new ；
S3: applying Pearson' S correlation coefficient method and recursive feature elimination method to the new feature data set D in step S2 new Carrying out feature screening, reserving the common features obtained by screening the two methods, and forming a feature data subset D sub ；
S4: subset D of feature data sub Dividing the training set and the test set, and carrying out normalization processing;
s5: constructing a Gaussian regression model;
s6: using the training set data for model training, outputting the model if the condition is met, otherwise, repeating S4-S6 until the model meets the condition;
s7: during the operation of the energy storage power station, when a certain battery pack exits the discharging operation, the current and the terminal voltage of the battery pack are collected, the voltage data of the whole process that the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharging is stopped, the instantaneous rise value of the terminal voltage at the discharging stopping moment and the terminal voltage after the discharging is stopped are recovered to the steady state value are collected, and the characteristics of the step S3 are extracted from the collected data by adopting the method of the step S2Data subset D sub And inputting the corresponding characteristic quantity into a trained Gaussian regression model, and estimating the SOH of the battery pack.
2. The method for estimating the SOH of the series-connected battery packs of the energy storage power stations as claimed in claim 1, wherein the original feature data set D of step S1 ori The method comprises the following specific construction steps:
s1-1: the allowable current interval [0, I ] of the battery pack max ]Performing equal difference division for N times, and obtaining a current data set comprising N current values as follows:
I＝[I 1 ,I 2 ,...,I N ]
s1-2: recording the number of single batteries contained in the battery pack as K, and setting the charging and discharging cut-off voltage range [ U ] of the battery pack pack-dis ，U pack-cha ]Performing M-order arithmetic division as voltage values of the suspended discharge point, and obtaining a voltage data set including M voltage values as follows:
U＝[U 1 ,U 2 ,...,U M ]
s1-3: from the current data set I ═ I 1 ,I 2 ,...,I N ]The method comprises the steps of selecting the ith current value to perform constant current discharge on a battery pack, pausing discharge for M times in the period, collecting discharge current and terminal voltage of the battery pack when the discharge is paused each time, collecting the terminal voltage of all battery monomers in the battery pack at the sampling moment before the discharge is paused, the instantaneous rise value of the terminal voltage at the discharge pausing moment, and the voltage data of the whole process that the terminal voltage is recovered to the steady state value after the discharge pausing, calculating SOH of the battery pack according to the total discharge when the discharge cutoff voltage is reached, constructing original data based on the obtained data, and regarding the mth pause discharge original data of the ith current, wherein the form is as follows:
D ori,i,m ＝[SOH i ,I i ,U i,m ,f Ucell-stop,i,m ,f ΔUcell,i,m ,f Ucell-re,i,m ,f Ucell-end,i,m ]
s1-4: step S1-3 is executed in a loop until the voltage data set U ═ U during the discharge of the battery pack at the ith current is acquired 1 ,U 2 ,…,U M ]The original data corresponding to all M voltage values in the data set constitute the following original data subset D ori,i ：
S1-5: from the current data set I ═ I 1 ,I 2 ,...,I N ]Sequentially selecting all current values, circularly executing the steps S1-3 to S1-4, and obtaining the original data of the battery pack in N multiplied by M times of discharge pause to form the following original data set D ori ：
3. The method of claim 1, wherein the new characteristic data set D of step S2 is used as the SOH estimation method for the series battery of energy storage power stations new The method comprises the following specific steps:
s2-1: extracting the range of the terminal voltage value at the sampling moment before the discharge of all the single batteries is suspended as a characteristic according to the m-th suspension discharge original data of the battery pack at the ith current, and marking the range as F Ucell-stop,i,m ；
S2-2: calculating the standard deviation of the terminal voltage value at the sampling moment before the discharge of all the single batteries is suspended according to the m-th suspended discharge original data of the battery pack at the ith current, and recording the standard deviation as std (f) Ucell-stop,i,m )；
S2-3: calculating the standard deviation of the instantaneous terminal voltage rise value at the discharge pause time of all the single batteries according to the m-th discharge pause original data of the battery pack at the ith current, and recording the standard deviation as std (f) ΔUcell,i,m )；
S2-4: calculating the whole process voltage recovery rate of the voltage recovery to a steady state value after the discharge of all the single batteries is suspended according to the m times of suspended discharge original data of the battery pack at the ith current, and recording the whole process voltage recovery rate as r Ucell-re,i,m ；
S2-5：Calculating the standard deviation of the steady-state terminal voltage values of all the single batteries as the characteristic according to the m-th pause discharge original data of the battery pack at the ith current, and recording the standard deviation as std (f) Ucell-end,i,m )；
S2-6: the features calculated in steps S2-1 to S2-5 are combined with the original data D of step S1-3 ori,i,m And combining to obtain a new characteristic data form of the battery pack in the m-th suspended discharge at the ith current as follows:
D new,i,m ＝[SOH i ,I i ,U i,m ,f Ucell-stop,i,m ,f ΔUcell,i,m ,f Ucell-re,i,m ,f Ucell-end,i,m ,F Ucell-stop,i,m ,std(f Ucell-stop,i,m ),std(f ΔUcell,i,m ),r Ucell-re,i,m ,std(f Ucell-end,i,m )]
s2-7: steps S2-1 to S2-6 are cyclically executed until all the new characteristic data of the battery pack discharged M times at the ith current value are acquired, and the following new characteristic data subset D is formed new,i ：
S2-8: from the current data set I ═ I 1 ,I 2 ,...,I N ]Sequentially selecting all current values, circularly executing the steps S2-1 to S2-7, and obtaining new characteristic data of the battery pack for N multiplied by M times of suspended discharge to form a new characteristic data set D new ：
4. The method for estimating the SOH of the series battery of the energy storage power station as claimed in claim 1, wherein the characteristic screening method of step S3 comprises the following steps:
s3-1: the new feature data set D obtained in step S2 new Each characteristic of (a) in turn makes a pearson correlation with the SOH of the battery packNumber inspection, retaining the feature with the absolute value of the number of the phase relation more than 0.8, removing the rest of the features to obtain a feature subset D sub-1 The specific calculation formula of the correlation coefficient is as follows:
wherein, the value range of r is [ -1, 1], when r is more than 0, the variables are linearly and positively correlated; when r < 0, the variables are linearly inversely correlated; when r is 0, there is no linear relationship between the variables.
S3-2: the new feature data set D obtained in step S2 new Inputting the initial feature subset into a random forest model, calculating the importance of each feature, calculating the estimation accuracy of the initial feature subset by using a cross-validation method, and removing the feature with the minimum importance from the initial feature subset to obtain a new feature subset; the operation is circulated to iterate until all the features are traversed, and according to the sequence of feature screening, the first 6 features are selected as high-quality features to form a feature subset D sub-2 ；
S3-3: feature subset D determined by steps S3-1 and S3-2 sub-1 And D sub-2 The common feature is preserved and constitutes the optimal feature subset D sub 。
5. The method for estimating the SOH of the series-connected battery packs of the energy storage power station as claimed in claim 1, wherein the SOH of one of the battery packs is estimated when the battery pack exits from the discharging operation in step S7, and the method comprises the following specific steps:
s7-1: importing the qualified Gaussian regression model trained in the step S6 and the kernel function parameters thereof into a battery pack management system of the energy storage power station;
s7-2: the battery pack management system of the energy storage power station collects and monitors the voltage and the current of all battery packs and all battery monomers in the battery packs in real time, when a certain battery pack exits from discharging operation, the current and the terminal voltage of the battery pack are collected, and the terminal voltage and the pause of the terminal voltage at the sampling moment before the time of the pause of the discharge of all the battery monomers in the battery pack are collectedThe voltage data of the whole process that the voltage at the discharging moment rises instantly and the voltage recovers to the steady state value after the discharging is suspended is adopted, and the characteristic data subset D of the step S3 is extracted from the collected data by the method of the step S2 sub A corresponding characteristic quantity;
s7-3: for the feature data subset D obtained in step S7-2 sub The result of the normalization process is input to the gaussian regression model introduced in step S7-1, and is estimated, thereby obtaining the SOH of the battery pack.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210435672.XA CN114910793B (en) | 2022-04-24 | 2022-04-24 | SOH estimation method for series battery pack of energy storage power station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210435672.XA CN114910793B (en) | 2022-04-24 | 2022-04-24 | SOH estimation method for series battery pack of energy storage power station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114910793A true CN114910793A (en) | 2022-08-16 |
| CN114910793B CN114910793B (en) | 2023-03-31 |
Family
ID=82764772
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210435672.XA Active CN114910793B (en) | 2022-04-24 | 2022-04-24 | SOH estimation method for series battery pack of energy storage power station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114910793B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110703112A (en) * | 2019-10-14 | 2020-01-17 | 重庆大学 | Online estimation method of battery pack state based on local charging data |
| CN111398837A (en) * | 2020-04-01 | 2020-07-10 | 重庆大学 | Vehicle battery health state estimation method based on data driving |
| WO2020182276A1 (en) * | 2019-03-08 | 2020-09-17 | Hochschule Anhalt | Method for estimating the state of charge soc of a battery assembly, corresponding apparatus for estimating the state of charge of a battery assembly and corresponding computer program |
| CN113092933A (en) * | 2021-03-23 | 2021-07-09 | 武汉大学 | LSTM-based single-phase earth fault line selection method and system |
| CN113189490A (en) * | 2021-05-17 | 2021-07-30 | 武汉理工大学 | Lithium battery health state estimation method based on feature screening and Gaussian process regression |
| CN113267733A (en) * | 2021-04-13 | 2021-08-17 | 西安理工大学 | Automatic configuration method for lithium battery health state estimation based on Gaussian process regression |
| CN113406521A (en) * | 2021-06-11 | 2021-09-17 | 上海电力大学 | Lithium battery health state online estimation method based on feature analysis |
| WO2021185308A1 (en) * | 2020-03-18 | 2021-09-23 | 北京理工大学 | Online determination method and system for state of health of power battery pack of electric vehicle |
| US20210311129A1 (en) * | 2020-04-03 | 2021-10-07 | Cummins Inc. | Methods and systems for accelerated determining of state of health using incremental capacity analysis |
| CN113552490A (en) * | 2021-06-29 | 2021-10-26 | 广东工业大学 | Reconfigurable battery pack SOC estimation method based on rest recovery effect |
| CN113552491A (en) * | 2021-06-29 | 2021-10-26 | 广东工业大学 | Energy storage power station battery pack SOC estimation method based on RNN model |
| US20210364574A1 (en) * | 2020-11-06 | 2021-11-25 | Beijing Institute Of Technology | Intelligent battery and state-of-charge online estimation method and applications thereof |
| CN113740736A (en) * | 2021-08-31 | 2021-12-03 | 哈尔滨工业大学 | Electric vehicle lithium battery SOH estimation method based on deep network self-adaptation |
| WO2021259196A1 (en) * | 2020-06-22 | 2021-12-30 | 北京理工大学 | Battery pack consistency evaluation method and system |
| CN113868597A (en) * | 2021-09-27 | 2021-12-31 | 电子科技大学 | Regression fairness measurement method for age estimation |
| CN113917334A (en) * | 2021-09-01 | 2022-01-11 | 西安理工大学 | Battery health state estimation method based on evolution LSTM self-encoder |
| CN114200333A (en) * | 2021-10-25 | 2022-03-18 | 国网天津市电力公司电力科学研究院 | Lead-acid storage battery capacity prediction method |
| CN114325393A (en) * | 2021-12-28 | 2022-04-12 | 江苏大学 | Lithium ion battery pack SOH self-adaptive estimation method based on PF and GPR |
-
2022
- 2022-04-24 CN CN202210435672.XA patent/CN114910793B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020182276A1 (en) * | 2019-03-08 | 2020-09-17 | Hochschule Anhalt | Method for estimating the state of charge soc of a battery assembly, corresponding apparatus for estimating the state of charge of a battery assembly and corresponding computer program |
| CN110703112A (en) * | 2019-10-14 | 2020-01-17 | 重庆大学 | Online estimation method of battery pack state based on local charging data |
| WO2021185308A1 (en) * | 2020-03-18 | 2021-09-23 | 北京理工大学 | Online determination method and system for state of health of power battery pack of electric vehicle |
| CN111398837A (en) * | 2020-04-01 | 2020-07-10 | 重庆大学 | Vehicle battery health state estimation method based on data driving |
| US20210311129A1 (en) * | 2020-04-03 | 2021-10-07 | Cummins Inc. | Methods and systems for accelerated determining of state of health using incremental capacity analysis |
| WO2021259196A1 (en) * | 2020-06-22 | 2021-12-30 | 北京理工大学 | Battery pack consistency evaluation method and system |
| US20210364574A1 (en) * | 2020-11-06 | 2021-11-25 | Beijing Institute Of Technology | Intelligent battery and state-of-charge online estimation method and applications thereof |
| CN113092933A (en) * | 2021-03-23 | 2021-07-09 | 武汉大学 | LSTM-based single-phase earth fault line selection method and system |
| CN113267733A (en) * | 2021-04-13 | 2021-08-17 | 西安理工大学 | Automatic configuration method for lithium battery health state estimation based on Gaussian process regression |
| CN113189490A (en) * | 2021-05-17 | 2021-07-30 | 武汉理工大学 | Lithium battery health state estimation method based on feature screening and Gaussian process regression |
| CN113406521A (en) * | 2021-06-11 | 2021-09-17 | 上海电力大学 | Lithium battery health state online estimation method based on feature analysis |
| CN113552491A (en) * | 2021-06-29 | 2021-10-26 | 广东工业大学 | Energy storage power station battery pack SOC estimation method based on RNN model |
| CN113552490A (en) * | 2021-06-29 | 2021-10-26 | 广东工业大学 | Reconfigurable battery pack SOC estimation method based on rest recovery effect |
| CN113740736A (en) * | 2021-08-31 | 2021-12-03 | 哈尔滨工业大学 | Electric vehicle lithium battery SOH estimation method based on deep network self-adaptation |
| CN113917334A (en) * | 2021-09-01 | 2022-01-11 | 西安理工大学 | Battery health state estimation method based on evolution LSTM self-encoder |
| CN113868597A (en) * | 2021-09-27 | 2021-12-31 | 电子科技大学 | Regression fairness measurement method for age estimation |
| CN114200333A (en) * | 2021-10-25 | 2022-03-18 | 国网天津市电力公司电力科学研究院 | Lead-acid storage battery capacity prediction method |
| CN114325393A (en) * | 2021-12-28 | 2022-04-12 | 江苏大学 | Lithium ion battery pack SOH self-adaptive estimation method based on PF and GPR |
Non-Patent Citations (1)
| Title |
|---|
| CHENGHUI ZHANG, 等: "A Novel Battery State-of-Health Estimation Method for Hybrid Electric Vehicles", 《IEEE/ASME TRANSACTIONS ON MECHATRONICS》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN115598557B (en) * | 2022-08-26 | 2023-08-25 | 广东工业大学 | Lithium battery SOH estimation method based on constant-voltage charging current |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114910793B (en) | 2023-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107957562B (en) | Online prediction method for residual life of lithium ion battery | |
| CN110568361B (en) | Method for predicting health state of power battery | |
| CN111007417A (en) | Battery pack SOH and RUL prediction method and system based on inconsistency evaluation | |
| CN106842056B (en) | Power battery peak power estimation method based on two-step on-line intelligence optimization algorithm | |
| US10677848B2 (en) | Apparatus, circuit model, and method for battery modelling | |
| CN107024663A (en) | The lithium battery screening technique clustered based on charging curve feature KPCA | |
| Li et al. | A new parameter estimation algorithm for an electrical analogue battery model | |
| CN111064253A (en) | Battery health degree rapid evaluation method based on average discrete Frechet distance | |
| Huang et al. | Accurate Thevenin's circuit-based battery model parameter identification | |
| CN106291375A (en) | A kind of SOC estimation method based on cell degradation and device | |
| CN111487532B (en) | Retired battery screening method and system based on analytic hierarchy process and entropy method | |
| CN114910793B (en) | SOH estimation method for series battery pack of energy storage power station | |
| CN115327415A (en) | Lithium battery SOC estimation method based on limited memory recursive least square algorithm | |
| Lin et al. | Estimating the state of health of lithium-ion batteries based on a probability density function | |
| CN116736172A (en) | Lithium battery pack health state evaluation method based on generation type countermeasure network | |
| Xiao et al. | State of health estimation framework of li-on battery based on improved Gaussian process regression for real car data | |
| CN116953547A (en) | Energy storage battery health evaluation method, system, equipment and storage medium | |
| CN114720879A (en) | Energy storage lithium battery pack aging mode automatic identification method based on BP neural network | |
| CN110639844B (en) | Lithium ion storage battery consistency screening method for space aircraft | |
| CN114252797A (en) | Uncertainty estimation-based lithium battery remaining service life prediction method | |
| CN114764124A (en) | Lithium battery SOC estimation method based on GAN and LSTM | |
| CN113300016A (en) | Battery charging and discharging control method and device | |
| CN112611973A (en) | Battery state of health estimation method, system, equipment and storage medium | |
| CN112649735A (en) | Method for rapidly obtaining total capacity of battery pack | |
| CN117368757B (en) | Performance detection method of needle type rechargeable 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 |