CN114035098A - Lithium battery health state prediction method integrating future working condition information and historical state information - Google Patents

Lithium battery health state prediction method integrating future working condition information and historical state information Download PDF

Info

Publication number
CN114035098A
CN114035098A CN202111525051.2A CN202111525051A CN114035098A CN 114035098 A CN114035098 A CN 114035098A CN 202111525051 A CN202111525051 A CN 202111525051A CN 114035098 A CN114035098 A CN 114035098A
Authority
CN
China
Prior art keywords
battery
data
model
sequence
health state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111525051.2A
Other languages
Chinese (zh)
Inventor
钱诚
徐炳辉
任羿
孙博
王自力
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN202111525051.2A priority Critical patent/CN114035098A/en
Publication of CN114035098A publication Critical patent/CN114035098A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/392—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]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • G01R31/3842—Arrangements for monitoring battery or accumulator variables, e.g. SoC combining voltage and current measurements

Abstract

The invention discloses a lithium battery health state prediction method integrating future working condition information and historical state information. Firstly, based on the cycle experiment data of the experimental battery under different working conditions, an experiment data set consisting of battery historical state data, data corresponding to future working conditions and data corresponding to future health states is obtained through processing. And then, constructing a multi-source sequence-to-sequence neural network model based on the attention layer, and training the model based on a battery experimental data set. And for the battery to be predicted, inputting historical health state data and future working condition data of the battery to be predicted into the trained multi-source sequence to the sequence model to obtain a health state prediction result of the battery. In addition, the model output is used as a new round of historical health state data, and the model is input iteratively to obtain a long-term prediction result. The method can be suitable for predicting the future health state of the lithium battery under different working conditions, has wide application conditions and high prediction precision, and has very high practical application value.

Description

Lithium battery health state prediction method integrating future working condition information and historical state information
Technical Field
The invention provides a lithium battery health state prediction method fusing future working condition information and historical state information, and belongs to the technical field of fault Prediction and Health Management (PHM).
Background
In recent years, with the continuous development of lithium battery technology, lithium batteries are being widely applied to various fields, such as electric vehicles, smart grids, and the like. However, during use, the state of health (SOH) of a lithium battery is gradually degraded due to various reasons such as precipitation of lithium metal, growth of a solid electrolyte membrane, and consumption of an electrolyte. When such degradation builds to a critical value, it can result in equipment that is inoperable, shut down, and even catastrophic safety issues. Therefore, there is a need for an accurate prediction of the SOH of a lithium battery to determine the maintenance or replacement time of the battery, thereby reducing the loss due to battery degradation. On the other hand, predicting the SOH of the lithium battery is also helpful for a battery management system to make a more reasonable battery balancing strategy, and the service life of the battery is prolonged.
The common lithium battery SOH prediction method mainly comprises two main categories, namely a model-based method and a data-driven method. Specifically, the model-based method generally describes the degradation behavior of the battery by using an electrochemical model, an equivalent circuit model or an empirical model, so as to construct a current SOH attenuation model, and estimates the parameters of the SOH attenuation model by combining a filtering algorithm and the like, thereby finally realizing the prediction of the SOH of the lithium battery. Unlike model-based methods, data-driven methods predict battery SOH based directly on battery historical SOH data rather than mathematical models. In the data-driven approach, commonly used algorithms include kernel-based approaches (e.g., support vector machines, correlation vector machines, etc.), neural network approaches (e.g., long-short term memory neural networks, etc.). However, the existing SOH prediction methods still have some disadvantages. For example, these methods only use historical state information of the battery as input, and ignore future operating condition information of the battery. Since the degradation of battery SOH is highly correlated with its operating conditions, ignoring future operating condition information will inevitably result in a loss of SOH prediction accuracy. Therefore, the above methods all have certain limitations in the application of lithium battery SOH prediction.
Disclosure of Invention
In order to overcome the defects of the existing lithium battery health state prediction technology, the invention provides the lithium battery health state prediction method integrating future working condition information and historical state information, which is suitable for predicting the future health states of the lithium battery under different working conditions and has high practical application value.
A health state prediction method fusing future working condition information and historical state information specifically comprises the following steps:
the first step is as follows: carrying out cycle experiments on the experimental battery under different working conditions, and recording time, battery terminal voltage and current data in the experimental process;
the second step is that: based on the battery experimental data, the battery state of health at each discharge cycle of the battery was calculated from the following formula:
Figure BDA0003408775290000021
wherein I (T) is battery discharge current, TkTotal discharge time of the battery at cycle k, CaprateFor nominal capacity, SOH, of the batterykIs the state of health of the battery at cycle k;
the third step: based on battery experimental data, constructing and calculating characteristic quantity describing battery working condition of each discharge cycle, and recording the working condition characteristic quantity of the battery in cycle k as lk；
The fourth step: with lwFor window length, use lsSliding to select experimental battery health state data for window step length
Figure BDA0003408775290000022
And the future l corresponding to each windowfCharacteristic data of operating conditions of each cycle
Figure BDA0003408775290000023
Selecting future health status data for the sample
Figure BDA0003408775290000024
Obtaining an experimental data set containing a plurality of samples as labels corresponding to the samples;
the fifth step: constructing a multi-source sequence-to-sequence model with 2 encoders and one decoderThe encoder 1, the encoder 2 and the decoder adopt network design of multilayer attention layers, and the encoder 1 and the encoder 2 respectively use battery health status data
Figure BDA0003408775290000025
Characteristic data of future working condition
Figure BDA0003408775290000026
Are inputs and all output hidden states to a decoder, which outputs a predicted battery health state;
and a sixth step: in reference to experimental data set of battery
Figure BDA0003408775290000031
In order to input the model, the model is input,
Figure BDA0003408775290000032
and adjusting the connection weight value of the sequence to the sequence model layer by layer from the output layer along the error reduction direction by adopting a reverse conduction algorithm for expecting the output of the model corresponding to the input until the accuracy of the model is converged to the optimal level, and finishing the training of the model.
The seventh step: for the battery to be predicted, the historical health state data and the working condition data of future circulation are taken as the input of the multi-source sequence trained in the sixth step to the sequence model, the obtained model output is the health state prediction result, and optionally, the model output can be used as the historical health state data to be iteratively input into the multi-source sequence to the sequence model to obtain the long-term health state prediction result.
Drawings
Fig. 1 is a flowchart illustrating steps of predicting the health status of a lithium battery in consideration of future operating conditions according to an embodiment of the present invention.
Fig. 2 is an actual state of health degradation curve of an experimental battery and a battery to be tested according to an embodiment of the present invention.
FIG. 3 is a multi-source sequence-to-sequence neural network model constructed in an embodiment of the present invention.
Fig. 4 shows the estimated result of the state of health of the battery to be tested according to the embodiment of the invention.
Detailed Description
The lithium battery health state prediction method fusing future working condition information and historical state information, which is provided by the invention, is further explained by combining the description of the attached drawings and the specific embodiment.
As shown in fig. 1, a method for predicting the health status of a lithium battery considering future working conditions includes the following steps:
s1, in the embodiment, the random charge and discharge cycle experimental data of six lithium batteries with the numbers of RW14, RW15, RW16, RW17, RW18 and RW19, which are provided by the national aerospace administration advanced prediction center (NASA PCoE), wherein four batteries in total of RW15, RW16, RW18 and RW19 are used as experimental batteries, and RW14 and RW17 are used as batteries to be predicted in terms of health states. To simulate the conditions in actual use, the cells were cycled to 4.2V in a cycling experiment and then discharged to 3.2V using a random discharge current between 0.5A and 5A, where the discharge current was updated at random samples per minute with a probability distribution as shown in table 1, which is referred to as a Random Walk (RW) cycle. After every 50RW cycles, one standard charge (i.e., constant current charge-constant voltage charge mode, CC-CV) and discharge (i.e., fixed current mode) cycle was performed to observe the state of health of the battery and calculate its SOH. In addition, a constant power discharge cycle of 15W and a pulsed current discharge cycle were performed after every 50RW cycle to observe changes in transient dynamics of the battery. During the whole aging process, all charging cycles of the battery follow a CC-CV mode, with a current of 2A during constant current charging and an off current of 0.01A during constant voltage charging. In a pulse current discharge cycle, the cell was discharged for 10 minutes at 1A and then left for 20 minutes, with discharge cycled and left until the voltage dropped to 3.2V.
TABLE 1 probability distribution of random walk discharge current of battery
Figure BDA0003408775290000041
S2, based on battery experimental data, calculating the battery health state of the battery in each discharge cycle according to the following formula:
Figure BDA0003408775290000042
wherein I (T) is battery discharge current, TkTotal discharge time of the battery at cycle k, Caprate is the nominal capacity of the battery, SOHkThe state of health of the battery in cycle k, in this example, the state of health degradation curves of the six batteries are finally obtained as shown in fig. 2.
The third step: based on the battery experimental data, a characteristic quantity describing the battery condition of each cycle is constructed and calculated, in the present embodiment, with the battery discharge current at [0A,5A ]]Within the interval, the discrete probability with 0.5A as the interval is taken as the characteristic quantity for describing the working condition of the battery and is marked as lk；
The fourth step: with lwFor window length, use lsSliding to select experimental battery health state data for window step length
Figure BDA0003408775290000051
And the future l corresponding to each windowfCharacteristic data of operating conditions of each cycle
Figure BDA0003408775290000052
Selecting future health status data for the input samples
Figure BDA0003408775290000053
As the output sample corresponding to the input sample, in the present embodiment, l is takenw＝50，ls＝1，lfThe experimental data set was finally obtained as 10.
The fifth step: constructing a multi-source sequence-to-sequence model with 2 encoders and one decoder, wherein the encoders and the decoders both adopt a multi-layer network design, and the encoders 1 and 2 respectively adopt a multi-layer network design
Figure BDA0003408775290000054
Are inputs and all output hidden states to a decoder which outputs the predicted battery state of health. In this implementationIn an example, a multi-source sequence-to-sequence model is constructed as shown in fig. 3, where both the encoder and decoder are composed of two attention layers in the form of residual concatenation.
And a sixth step: in reference to experimental data set of battery
Figure BDA0003408775290000055
In order to input the model, the model is input,
Figure BDA0003408775290000056
and adjusting the connection weight from the sequence to the sequence model layer by layer for corresponding output of the model by adopting a reverse conduction algorithm along the direction of error reduction until the accuracy of the model is not improved any more, and finishing model training by the model.
The seventh step: for the batteries RW14 and RW17 to be predicted, historical health state data and future cycle working condition data of the batteries RW14 and RW17 are respectively taken as input of the multi-source sequence to the sequence model trained in the sixth step, obtained model output is a health state prediction result, the model output is taken as historical health state data, the multi-source sequence is input in an iteration mode to the sequence model, and finally long-term health state prediction results of the batteries RW14 and RW17 are obtained and are shown in fig. 4. The root mean square error between the predicted value and the actual value of the health states of the batteries RW14 and RW17 is 1.76% and 1.84%, which shows the accuracy of the prediction result and proves the feasibility and the effectiveness of the method in the lithium battery health state prediction.

Claims (3)

1. A lithium battery health state prediction method integrating future working condition information and historical state information is characterized by comprising the following steps:
step 1: carrying out cyclic experiments on the experimental battery under different working conditions, and recording time, battery voltage and current data in the experimental process;
step 2: analyzing and processing the voltage, current and time data of the experimental battery to form an experimental data set;
and step 3: constructing a multi-source sequence-to-sequence neural network model taking historical health state data and future working condition data as input, and training the model based on an experimental data set to obtain a trained multi-source sequence-to-sequence model;
and 4, step 4: and (3) for the battery to be predicted, taking historical health state data and future working condition data of the battery as input of the multi-source sequence trained in the step (3) to the sequence model, and obtaining model output which is a health state prediction result.
2. The method for predicting the health status of a lithium battery as claimed in claim 1, wherein the step 2 further comprises the following steps:
step 21: based on the battery experimental data, the battery state of health at each discharge cycle of the battery was calculated from the following formula:
Figure FDA0003408775280000011
wherein I (T) is battery discharge current, TkTotal discharge time of the battery at cycle k, CaprateFor nominal capacity, SOH, of the batterykIs the state of health of the battery at cycle k;
step 22: based on battery experimental data, constructing and calculating a battery working condition characteristic quantity for describing each discharge cycle, and recording the working condition characteristic quantity of the battery in cycle k as lk；
Step 23: with lwFor window length, use lsSliding to select experimental battery health state data for window step length
Figure FDA0003408775280000012
And the future l corresponding to each windowfCharacteristic data of operating conditions of each cycle
Figure FDA0003408775280000021
Selecting future health status data for the sample
Figure FDA0003408775280000022
As a label corresponding to a sample, an experimental data set containing several samples is obtained.
3. The method for predicting the health status of a lithium battery as claimed in claim 1, wherein the step 3 further comprises the following steps:
step 31: constructing a multi-source sequence-to-sequence model with 2 encoders and 1 decoder, wherein each encoder 1, each encoder 2 and each decoder are composed of attention layers, and each encoder 1 and each encoder 2 respectively use battery health state data
Figure FDA0003408775280000027
Characteristic data of future working condition
Figure FDA0003408775280000026
Are inputs and all output hidden states to a decoder, which outputs a predicted battery health state;
step 32: in reference to experimental data set of battery
Figure FDA0003408775280000023
Figure FDA0003408775280000025
In order to input the model, the model is input,
Figure FDA0003408775280000024
and adjusting the connection weight value of the sequence to the sequence model layer by layer from the output layer along the error reduction direction by adopting a reverse conduction algorithm for expecting the output of the model corresponding to the input until the accuracy of the model is converged to the optimal level, and finishing the training of the model.
CN202111525051.2A 2021-12-14 2021-12-14 Lithium battery health state prediction method integrating future working condition information and historical state information Pending CN114035098A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111525051.2A CN114035098A (en) 2021-12-14 2021-12-14 Lithium battery health state prediction method integrating future working condition information and historical state information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111525051.2A CN114035098A (en) 2021-12-14 2021-12-14 Lithium battery health state prediction method integrating future working condition information and historical state information

Publications (1)

Publication Number Publication Date
CN114035098A true CN114035098A (en) 2022-02-11

Family

ID=80140471

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111525051.2A Pending CN114035098A (en) 2021-12-14 2021-12-14 Lithium battery health state prediction method integrating future working condition information and historical state information

Country Status (1)

Country Link
CN (1) CN114035098A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115291108A (en) * 2022-06-27 2022-11-04 东莞新能安科技有限公司 Data generation method, device, equipment and computer program product
CN116148700A (en) * 2023-01-10 2023-05-23 中国第一汽车股份有限公司 Method for predicting state of battery and storage medium
CN116430244A (en) * 2023-06-14 2023-07-14 聊城大学 Power battery health state estimation method based on voltage and current characteristics
EP4270033A1 (en) * 2022-04-28 2023-11-01 Samsung SDI Co., Ltd. Method and apparatus for estimating state of health of battery
CN117269766A (en) * 2023-08-28 2023-12-22 广东工业大学 Battery SOH prediction method for unbalanced use scene
CN116148700B (en) * 2023-01-10 2024-04-12 中国第一汽车股份有限公司 Method for predicting state of battery and storage medium

Citations (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103273921A (en) * 2013-06-14 2013-09-04 清华大学 Method for estimating driving range of electric car
CN103649858A (en) * 2011-05-31 2014-03-19 空中客车运营有限公司 Method and device for predicting the condition of a component or system, computer program product
CN105678340A (en) * 2016-01-20 2016-06-15 福州大学 Automatic image marking method based on enhanced stack type automatic encoder
US20160363629A1 (en) * 2015-06-12 2016-12-15 GM Global Technology Operations LLC Systems and methods for estimating battery system power capability
CN106483470A (en) * 2016-12-22 2017-03-08 清华大学 Battery residual discharge energy prediction method based on future operation condition prediction
CN107194438A (en) * 2017-05-24 2017-09-22 武汉大学 A kind of depth characteristic method for expressing based on multiple stack own coding
CN107696896A (en) * 2017-09-29 2018-02-16 江西江铃集团新能源汽车有限公司 Electric automobile continual mileage evaluation method
US20180080995A1 (en) * 2016-09-20 2018-03-22 Faraday&Future Inc. Notification system and method for providing remaining running time of a battery
US20180143257A1 (en) * 2016-11-21 2018-05-24 Battelle Energy Alliance, Llc Systems and methods for estimation and prediction of battery health and performance
CN108445410A (en) * 2018-04-02 2018-08-24 国家计算机网络与信息安全管理中心 A kind of method and device of monitoring accumulator group operating status
CN108776308A (en) * 2018-04-26 2018-11-09 北京长城华冠汽车科技股份有限公司 Estimate the method and device of battery remaining power
CN109870659A (en) * 2019-03-14 2019-06-11 燕山大学 Using the health state of lithium ion battery evaluation method of sliding window optimizing strategy
WO2019184841A1 (en) * 2018-03-30 2019-10-03 比亚迪股份有限公司 Electric vehicle, and management system and method for power battery therein
CN110378052A (en) * 2019-07-25 2019-10-25 北京航空航天大学 It is looked to the future the equipment method for predicting residual useful life of operating condition based on Recognition with Recurrent Neural Network
CN111191824A (en) * 2019-12-20 2020-05-22 北京理工新源信息科技有限公司 Power battery capacity attenuation prediction method and system
CN111832220A (en) * 2020-06-16 2020-10-27 天津大学 Lithium ion battery health state estimation method based on codec model
CN111860785A (en) * 2020-07-24 2020-10-30 中山大学 Time sequence prediction method and system based on attention mechanism cyclic neural network
CN111999657A (en) * 2020-10-29 2020-11-27 北京航空航天大学 Method for evaluating driving mileage of lithium ion battery of electric vehicle in residual life
WO2021000362A1 (en) * 2019-07-04 2021-01-07 浙江大学 Deep neural network model-based address information feature extraction method
CN112782591A (en) * 2021-03-22 2021-05-11 浙江大学 Lithium battery SOH long-term prediction method based on multi-battery data fusion
CN112937547A (en) * 2021-01-28 2021-06-11 北京理工大学 Plug-in hybrid power bus energy management method based on global working conditions
CN113138342A (en) * 2021-03-17 2021-07-20 江苏大学 SOC online estimation method and system based on rolling time domain estimation
CN113419187A (en) * 2021-06-08 2021-09-21 上海交通大学 Lithium ion battery health estimation method
CN113434970A (en) * 2021-06-01 2021-09-24 北京交通大学 Health index curve extraction and service life prediction method for mechanical equipment
WO2021244632A1 (en) * 2020-06-05 2021-12-09 北京理工大学 Electric automobile energy consumption prediction method and system

Patent Citations (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103649858A (en) * 2011-05-31 2014-03-19 空中客车运营有限公司 Method and device for predicting the condition of a component or system, computer program product
CN103273921A (en) * 2013-06-14 2013-09-04 清华大学 Method for estimating driving range of electric car
US20160363629A1 (en) * 2015-06-12 2016-12-15 GM Global Technology Operations LLC Systems and methods for estimating battery system power capability
CN105678340A (en) * 2016-01-20 2016-06-15 福州大学 Automatic image marking method based on enhanced stack type automatic encoder
US20180080995A1 (en) * 2016-09-20 2018-03-22 Faraday&Future Inc. Notification system and method for providing remaining running time of a battery
US20180143257A1 (en) * 2016-11-21 2018-05-24 Battelle Energy Alliance, Llc Systems and methods for estimation and prediction of battery health and performance
CN106483470A (en) * 2016-12-22 2017-03-08 清华大学 Battery residual discharge energy prediction method based on future operation condition prediction
CN107194438A (en) * 2017-05-24 2017-09-22 武汉大学 A kind of depth characteristic method for expressing based on multiple stack own coding
CN107696896A (en) * 2017-09-29 2018-02-16 江西江铃集团新能源汽车有限公司 Electric automobile continual mileage evaluation method
WO2019184841A1 (en) * 2018-03-30 2019-10-03 比亚迪股份有限公司 Electric vehicle, and management system and method for power battery therein
CN108445410A (en) * 2018-04-02 2018-08-24 国家计算机网络与信息安全管理中心 A kind of method and device of monitoring accumulator group operating status
CN108776308A (en) * 2018-04-26 2018-11-09 北京长城华冠汽车科技股份有限公司 Estimate the method and device of battery remaining power
CN109870659A (en) * 2019-03-14 2019-06-11 燕山大学 Using the health state of lithium ion battery evaluation method of sliding window optimizing strategy
WO2021000362A1 (en) * 2019-07-04 2021-01-07 浙江大学 Deep neural network model-based address information feature extraction method
CN110378052A (en) * 2019-07-25 2019-10-25 北京航空航天大学 It is looked to the future the equipment method for predicting residual useful life of operating condition based on Recognition with Recurrent Neural Network
CN111191824A (en) * 2019-12-20 2020-05-22 北京理工新源信息科技有限公司 Power battery capacity attenuation prediction method and system
WO2021244632A1 (en) * 2020-06-05 2021-12-09 北京理工大学 Electric automobile energy consumption prediction method and system
CN111832220A (en) * 2020-06-16 2020-10-27 天津大学 Lithium ion battery health state estimation method based on codec model
CN111860785A (en) * 2020-07-24 2020-10-30 中山大学 Time sequence prediction method and system based on attention mechanism cyclic neural network
CN111999657A (en) * 2020-10-29 2020-11-27 北京航空航天大学 Method for evaluating driving mileage of lithium ion battery of electric vehicle in residual life
CN112937547A (en) * 2021-01-28 2021-06-11 北京理工大学 Plug-in hybrid power bus energy management method based on global working conditions
CN113138342A (en) * 2021-03-17 2021-07-20 江苏大学 SOC online estimation method and system based on rolling time domain estimation
CN112782591A (en) * 2021-03-22 2021-05-11 浙江大学 Lithium battery SOH long-term prediction method based on multi-battery data fusion
CN113434970A (en) * 2021-06-01 2021-09-24 北京交通大学 Health index curve extraction and service life prediction method for mechanical equipment
CN113419187A (en) * 2021-06-08 2021-09-21 上海交通大学 Lithium ion battery health estimation method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
W.A. ADAMS等: "Cost/benefit analyses of a new battery pack management technique for telecommunication applications: future directions with fuel cell/battery systems", 《INTELEC 2004. 26TH ANNUAL INTERNATIONAL TELECOMMUNICATIONS ENERGY CONFERENCE》 *
李金东等: "退役锂离子电池健康状态评估方法综述", 《储能科学与技术》 *
赵显赫等: "基于数据驱动的锂离子电池健康状态评估综述", 《浙江电力》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4270033A1 (en) * 2022-04-28 2023-11-01 Samsung SDI Co., Ltd. Method and apparatus for estimating state of health of battery
CN115291108A (en) * 2022-06-27 2022-11-04 东莞新能安科技有限公司 Data generation method, device, equipment and computer program product
CN116148700A (en) * 2023-01-10 2023-05-23 中国第一汽车股份有限公司 Method for predicting state of battery and storage medium
CN116148700B (en) * 2023-01-10 2024-04-12 中国第一汽车股份有限公司 Method for predicting state of battery and storage medium
CN116430244A (en) * 2023-06-14 2023-07-14 聊城大学 Power battery health state estimation method based on voltage and current characteristics
CN116430244B (en) * 2023-06-14 2023-08-15 聊城大学 Power battery health state estimation method based on voltage and current characteristics
CN117269766A (en) * 2023-08-28 2023-12-22 广东工业大学 Battery SOH prediction method for unbalanced use scene

Similar Documents

Publication Publication Date Title
CN110187290B (en) Lithium ion battery residual life prediction method based on fusion algorithm
CN114035098A (en) Lithium battery health state prediction method integrating future working condition information and historical state information
CN110161425B (en) Method for predicting remaining service life based on lithium battery degradation stage division
CN105277896B (en) Lithium battery method for predicting residual useful life based on ELM MUKF
Chen et al. Prognostics of lithium-ion batteries using model-based and data-driven methods
Jiang et al. State of health estimation for lithium-ion battery using empirical degradation and error compensation models
CN114372417A (en) Electric vehicle battery health state and remaining life evaluation method based on charging network
CN104316879B (en) A kind of prediction technique in lead-acid batteries service life
CN111856287A (en) Lithium battery health state detection method based on stacked residual causal convolutional neural network
CN111753416A (en) Lithium ion battery RUL prediction method based on two-stage Wiener process
CN111999648A (en) Lithium battery residual life prediction method based on long-term and short-term memory network
CN115951254A (en) Lithium ion battery health state and remaining service life estimation method
CN111308375A (en) LSTM-FFNN-based electric forklift lithium ion battery health state prediction method
Vilsen et al. Log-linear model for predicting the lithium-ion battery age based on resistance extraction from dynamic aging profiles
Zou et al. Advancements in Artificial Neural Networks for health management of energy storage lithium-ion batteries: A comprehensive review
CN112937369B (en) Active equalization control method for power battery pack based on Mahalanobis process
CN113791351A (en) Lithium battery life prediction method based on transfer learning and difference probability distribution
CN116653608B (en) Electric automobile charging protection and control method, device and storage medium
CN112327169A (en) Lithium battery residual life prediction method
CN113721151B (en) Battery capacity estimation model and method based on double-tower deep learning network
CN116047314A (en) Rechargeable battery health state prediction method
Han et al. Integration of long-short term memory network and fuzzy logic for high-safety in a FR-ESS with degradation and failure
CN115621573A (en) System and method for predicting health state of lithium battery of energy storage power station
Udeogu et al. Remaining useful life prediction for supercapacitors using an optimized end-to-end deep learning approach
CN114896865A (en) Digital twin-oriented self-adaptive evolutionary neural network health state online prediction method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination