CN112798960B - A battery pack remaining life prediction method based on transfer deep learning - Google Patents
- ️Fri Jun 24 2022
CN112798960B - A battery pack remaining life prediction method based on transfer deep learning - Google Patents
A battery pack remaining life prediction method based on transfer deep learning Download PDFInfo
-
Publication number
- CN112798960B CN112798960B CN202110048627.4A CN202110048627A CN112798960B CN 112798960 B CN112798960 B CN 112798960B CN 202110048627 A CN202110048627 A CN 202110048627A CN 112798960 B CN112798960 B CN 112798960B Authority
- CN
- China Prior art keywords
- battery
- battery pack
- cell
- capacity
- model Prior art date
- 2021-01-14 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
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000013135 deep learning Methods 0.000 title claims abstract description 16
- 238000012546 transfer Methods 0.000 title description 5
- 230000036541 health Effects 0.000 claims abstract description 96
- 230000032683 aging Effects 0.000 claims abstract description 28
- 238000010801 machine learning Methods 0.000 claims abstract description 15
- 238000009826 distribution Methods 0.000 claims abstract description 12
- 238000013526 transfer learning Methods 0.000 claims abstract description 12
- 238000010219 correlation analysis Methods 0.000 claims abstract description 7
- 238000013508 migration Methods 0.000 claims abstract description 7
- 230000005012 migration Effects 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 7
- 230000032677 cell aging Effects 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims description 13
- 239000000178 monomer Substances 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 5
- 230000000875 corresponding effect Effects 0.000 claims description 5
- 238000013136 deep learning model Methods 0.000 claims description 5
- 238000002474 experimental method Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000002596 correlated effect Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 14
- 238000000605 extraction Methods 0.000 description 10
- 230000000694 effects Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007787 long-term memory Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Secondary Cells (AREA)
Abstract
本发明涉及一种基于迁移深度学习的电池组剩余寿命预测方法，属于电池技术领域。该方法包括以下步骤：步骤S1：收集动力电池老化数据集，建立电池老化数据库；步骤S2：根据电池单体老化数据提取多个健康因子，并根据相关性分析和容量估计误差筛选健康因子；步骤S3：基于电池单体全寿命周期老化数据集训练得到健康因子的递推模型，以及基于健康因子的容量估计模型；步骤S4：建立基于电池单体健康因子集和电池组容量衰减的机器学习模型；步骤S5：基于单体容量估计模型预测未来各单体的容量，得到未来循环的电池组单体容量分布。利用迁移学习和深度学习相结合，能够有效的利用已有的完整信息，提高电池组剩余寿命预测精度。
The invention relates to a method for predicting the remaining life of a battery pack based on migration deep learning, and belongs to the technical field of batteries. The method includes the following steps: step S1: collecting power battery aging data sets and establishing a battery aging database; step S2: extracting multiple health factors according to the battery cell aging data, and screening the health factors according to correlation analysis and capacity estimation error; step S3: The recursive model of the health factor and the capacity estimation model based on the health factor are trained based on the battery cell life cycle aging data set; Step S4: The machine learning model based on the battery cell health factor set and the capacity decay of the battery pack is established ; Step S5: Predict the capacity of each cell in the future based on the cell capacity estimation model, and obtain the cell capacity distribution of the battery pack in the future cycle. The combination of transfer learning and deep learning can effectively use the existing complete information to improve the prediction accuracy of the remaining life of the battery pack.
Description
技术领域technical field
本发明属于电池技术领域，涉及一种基于迁移深度学习的电池组剩余寿命预测方法。The invention belongs to the technical field of batteries, and relates to a method for predicting the remaining life of a battery pack based on migration deep learning.
背景技术Background technique
电池的剩余寿命预测方法通常可分为基于模型和基于数据驱动两类。基于模型的一种方法主要是通过历史数据与循环次数建立的经验或半经验模型进行预测。通常为指数模型、双指数模型、或者多项式模型。利用先进滤波器例如卡尔曼滤波，粒子滤波等进行曲线拟合得到拟合曲线，从而利用拟合曲线进行容量估计或者进行剩余寿命预测。基于物理模型是另外一种基于模型的方法，该方法建立电池的老化机理模型，从而通过仿真实现未来循环的充放电模拟，进而得到容量衰减至阈值时的剩余寿命。而基于数据驱动的方法由于其不需要特殊的模型，而仅仅依赖数据本身的特性，近年来得到了快速的发展。基于数据驱动的方法通常是以容量衰减的序列构建映射关系，通过先前几个容量数据预测下一个容量数据，并进行外推得到剩余循环寿命。或者通过在充放电过程中根据特殊的工况提取相关的健康因子，建立健康因子和容量之间的映射关系，通过健康因子估计电池的容量。然后建立电池容量序列映射关系进行外推得到电池的剩余循环寿命。然而，目前的研究尚缺乏对电池组的寿命预测的实现。迁移学习能够利用有效的历史信息，对预测任务进行模型的微小修正，从而提高预测任务的效果。目前研究仍然缺乏有效的利用电池单电芯和电池组的对应关系，从而对电池组的剩余寿命的预测精度提高。此外，电池组的剩余寿命不仅需要关注整包的未来容量变化，也需要关注每个电池单体的容量分布，从而可以识别出不一致性比较大的电池单体，及时更换或进行均衡管理，以延长电池组的使用寿命。The remaining life prediction methods of batteries can generally be divided into two categories: model-based and data-driven. A model-based method is mainly based on empirical or semi-empirical models established by historical data and cycle times for prediction. Usually an exponential model, a double exponential model, or a polynomial model. Use advanced filters such as Kalman filter, particle filter, etc. to perform curve fitting to obtain a fitted curve, so as to use the fitted curve for capacity estimation or remaining life prediction. Physical model-based is another model-based method, which establishes the aging mechanism model of the battery, so as to realize the charging and discharging simulation of future cycles through simulation, and then obtain the remaining life when the capacity decays to the threshold value. However, data-driven methods have developed rapidly in recent years because they do not require special models, but only rely on the characteristics of the data itself. Data-driven methods usually build a mapping relationship in the sequence of capacity decay, predict the next capacity data from the previous capacity data, and extrapolate to obtain the remaining cycle life. Or by extracting the relevant health factors according to special working conditions during the charging and discharging process, establishing the mapping relationship between the health factors and the capacity, and estimating the capacity of the battery through the health factors. Then the battery capacity sequence mapping relationship is established to extrapolate to obtain the remaining cycle life of the battery. However, the current research lacks the realization of the life prediction of the battery pack. Transfer learning can use effective historical information to make minor corrections to the prediction task model, thereby improving the effect of the prediction task. At present, there is still a lack of effective use of the corresponding relationship between battery cells and battery packs, so as to improve the prediction accuracy of the remaining life of the battery pack. In addition, the remaining life of the battery pack not only needs to pay attention to the future capacity changes of the whole package, but also needs to pay attention to the capacity distribution of each battery cell, so that the battery cells with relatively large inconsistencies can be identified and replaced in time or balanced management is carried out to avoid Extend the life of the battery pack.
针对上述的问题，目前尚未提出有效的基于迁移深度学习的电池组剩余寿命和电池组单体寿命分布的高效精确预测方法。In view of the above problems, an efficient and accurate prediction method based on transfer deep learning for the remaining battery life and battery unit life distribution has not yet been proposed.
发明内容SUMMARY OF THE INVENTION
有鉴于此，本发明的目的在于提供一种基于迁移深度学习的电池组剩余寿命预测方法。In view of this, the purpose of the present invention is to provide a method for predicting the remaining life of a battery pack based on transfer deep learning.
为达到上述目的，本发明提供如下技术方案：To achieve the above object, the present invention provides the following technical solutions:
一种基于迁移深度学习的电池组剩余寿命预测方法，该方法包括以下步骤：A method for predicting the remaining life of a battery pack based on transfer deep learning, the method includes the following steps:
步骤S1：收集动力电池老化数据集，建立电池老化数据库，包括完整的电池单体全寿命周期数据以及由同款电芯组成的电池组早期的老化数据；Step S1: collect power battery aging data sets, and establish a battery aging database, including complete battery life cycle data and early aging data of battery packs composed of cells of the same type;
步骤S2：根据电池单体老化数据提取多个健康因子，并根据相关性分析和容量估计误差筛选健康因子；Step S2: extracting multiple health factors according to battery cell aging data, and screening health factors according to correlation analysis and capacity estimation error;
步骤S3：基于电池单体全寿命周期老化数据集训练得到健康因子的递推模型，以及基于健康因子的容量估计模型；Step S3: training a recursive model of the health factor and a capacity estimation model based on the health factor based on the battery cell life cycle aging data set;
步骤S4：提取电池组早期老化实验数据集每个电池单体的健康因子，建立基于电池单体健康因子集和电池组容量衰减的机器学习模型；Step S4: extracting the health factor of each battery cell in the battery pack early aging experiment data set, and establishing a machine learning model based on the battery cell health factor set and the battery pack capacity decay;
步骤S5：利用迁移学习对电池组每个电池单体的健康因子递推模型进行微调，并进行健康因子外推预测，最后基于外推得到的单体健康因子集估计未来循环的电池组容量；基于单体容量估计模型预测未来各单体的容量，得到未来循环的电池组单体容量分布。Step S5: using transfer learning to fine-tune the recursive model of the health factor of each battery cell in the battery pack, and extrapolate the health factor to predict, and finally estimate the capacity of the battery pack in the future cycle based on the extrapolated cell health factor set; Based on the cell capacity estimation model, the capacity of each cell in the future is predicted, and the cell capacity distribution of the battery pack in the future cycle is obtained.
可选的，所述步骤S1具体为：Optionally, the step S1 is specifically:
步骤S11：收集某款电芯的多个电池单体的全寿命周期完整数据集，包括充放电电流、电压、温度、时间和电量参数，涵盖不同充放电倍率和环境温度的数据集；Step S11 : collect a complete data set of the full life cycle of a plurality of battery cells of a certain battery cell, including charge and discharge current, voltage, temperature, time and power parameters, and data sets covering different charge and discharge rates and ambient temperatures;
步骤S12：收集电池单体同款电芯对应的电池组的老化实验数据集，包括每个单体的电压和温度参数，以及电池组电流和电量信息；Step S12 : collecting the aging experiment data set of the battery pack corresponding to the same type of battery cell, including the voltage and temperature parameters of each cell, and the current and power information of the battery pack;
步骤S13：根据收集的电池单体和电池组的老化实验数据，建立某款电芯的电池老化数据集。Step S13 : establishing a battery aging data set of a certain type of battery cell according to the collected aging experimental data of the battery cell and the battery pack.
可选的，所述步骤S2具体为：Optionally, the step S2 is specifically:
步骤S21：根据电池单体全寿命周期提取多个可用的健康因子；Step S21: extracting a plurality of available health factors according to the full life cycle of the battery cell;
步骤S22：通过相关性分析，筛选出和电池单体容量相关性高的健康因子；Step S22: Screening out health factors that are highly correlated with the battery cell capacity through correlation analysis;
步骤S23：基于A号电池单体建立全寿命的健康因子估计容量的机器学习模型，并利用B号电池单体提取健康因子进行容量估计，分析估计误差。Step S23 : establishing a machine learning model for estimating the capacity by the health factor of the whole life based on the battery cell A, and using the battery cell B to extract the health factor to estimate the capacity, and analyze the estimation error.
可选的，所述多个可用的健康因子包括电压曲线斜率，等时间间隔电压/温度变化，等电压/温度间隔电量变化，电量序列方差，电量差方差，容量增量曲线峰值、谷值、峰间隔、峰面积、电压/温度差分曲线峰值、谷值和峰间隔；Optionally, the multiple available health factors include voltage curve slope, voltage/temperature change at equal time intervals, power change at equal voltage/temperature intervals, power sequence variance, power difference variance, peak, valley, Peak interval, peak area, voltage/temperature differential curve peak, valley and peak interval;
A号电池单体和B号电池单体指的是同款电芯的不同电池单体，机器学习模型为高斯过程回归模型或相关向量机模型，实现概率预测。A battery cell and a battery cell B refer to different battery cells of the same cell, and the machine learning model is a Gaussian process regression model or a correlation vector machine model to achieve probability prediction.
可选的，所述步骤S3具体为：Optionally, the step S3 is specifically:
步骤S31：根据选定的健康因子，建立多个单体的全寿命周期健康因子递推衰减的深度学习模型；Step S31: According to the selected health factor, establish a deep learning model of recursive decay of the life cycle health factor of multiple monomers;
步骤S32：根据选的的健康因子，建立多个单体的全寿命周期基于健康因子的容量估计模型。Step S32: According to the selected health factor, establish a capacity estimation model based on the health factor for the whole life cycle of a plurality of monomers.
可选的，所述深度学习模型具体指多层神经网络构建的模型，为高斯过程回归模型或相关向量机模型，实现概率预测。Optionally, the deep learning model specifically refers to a model constructed by a multi-layer neural network, which is a Gaussian process regression model or a correlation vector machine model, and implements probability prediction.
可选的，所述步骤S4具体为：Optionally, the step S4 is specifically:
步骤S41：根据选定的健康因子，提取电池组早期循环中每个电池单体的健康因子，建立电池组各单体的健康因子组成的特征集；Step S41: according to the selected health factor, extract the health factor of each battery cell in the early cycle of the battery pack, and establish a feature set composed of the health factor of each cell of the battery pack;
步骤S42：根据单体特征集和电池组容量真值，建立电池组容量估计的机器学习模型。Step S42: Establish a machine learning model for estimating battery pack capacity according to the cell feature set and the true value of the battery pack capacity.
可选的，所述特征集为电池组中每个电池单体特征组成的特征矩阵，为高斯过程回归模型或相关向量机模型，实现概率预测。Optionally, the feature set is a feature matrix composed of features of each battery cell in the battery pack, and is a Gaussian process regression model or a correlation vector machine model to implement probability prediction.
可选的，所述步骤S5具体为：Optionally, the step S5 is specifically:
步骤S51：利用前期训练好的电池单体全寿命周期健康因子衰减模型和电池组各单体早期提取的健康因子，运用迁移学习的方法将训练好的网络进行部分结构再训练，得到电池组每个单体的健康因子递推模型；Step S51: Using the pre-trained battery cell life cycle health factor decay model and the health factors extracted early from each cell of the battery pack, use the transfer learning method to retrain the trained network for part of the structure, and obtain the battery pack each time. A recursive model of the health factor of a single individual;
步骤S52：根据每个单体的健康因子递推模型，外推得到未来未知循环次数的健康因子预测值；Step S52: According to the recursive model of the health factor of each monomer, extrapolate to obtain the predicted value of the health factor of the unknown number of cycles in the future;
步骤S53：根据建立的电池单体特征集估计电池组容量的机器学习模型，预测未来循环的电池组容量值；Step S53: according to the established battery cell feature set, a machine learning model for estimating the capacity of the battery pack, and predicting the capacity value of the battery pack in the future cycle;
步骤S54：根据建立的电池单体容量估计模型，得到未来未知循环的电池组内各单体的容量预测值，从而得到电池组单体的容量预测分布；Step S54: According to the established battery cell capacity estimation model, obtain the capacity prediction value of each cell in the battery pack for unknown cycles in the future, so as to obtain the capacity prediction distribution of the battery pack cells;
迁移学习的方法将基于早期训练好的健康因子递推模型的深度学习网络，选择性的冻结某些层，对剩下的层进行模型再训练，或拆掉某些层并重新构建网络进行新网络的再训练。The transfer learning method will be based on the deep learning network of the early trained health factor recursive model, selectively freeze some layers, retrain the model for the remaining layers, or remove some layers and rebuild the network for new Retraining of the network.
本发明的有益效果在于：The beneficial effects of the present invention are:
1)利用迁移学习和深度学习相结合，能够有效的利用已有的完整信息，提高电池组剩余寿命预测精度。1) The combination of transfer learning and deep learning can effectively use the existing complete information to improve the prediction accuracy of the remaining life of the battery pack.
2)所提出的电池组剩余寿命预测可实现整组寿命预测和电池单体寿命分布预测。2) The proposed battery pack remaining life prediction can realize the whole group life prediction and the battery cell life distribution prediction.
3)所提出的方案可以指导电池组的设计和开发，仅利用早期数据进行寿命预测，从而优化电池组的设计。3) The proposed scheme can guide the design and development of the battery pack, using only early data for lifetime prediction, thereby optimizing the design of the battery pack.
4)所提出的方案提供了概率性预测，提供预测的置信分布。4) The proposed scheme provides probabilistic predictions, providing confidence distributions for the predictions.
本发明的其他优点、目标和特征在某种程度上将在随后的说明书中进行阐述，并且在某种程度上，基于对下文的考察研究对本领域技术人员而言将是显而易见的，或者可以从本发明的实践中得到教导。本发明的目标和其他优点可以通过下面的说明书来实现和获得。Other advantages, objects, and features of the present invention will be set forth in the description that follows, and will be apparent to those skilled in the art based on a study of the following, to the extent that is taught in the practice of the present invention. The objectives and other advantages of the present invention may be realized and attained by the following description.
附图说明Description of drawings
为了使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明作优选的详细描述，其中：In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be preferably described in detail below with reference to the accompanying drawings, wherein:
图1为本发明整体的方法流程图；Fig. 1 is the overall method flow chart of the present invention;
图2为实施例的健康因子示意图；(a)为基于测量数据特征提取示意图；(b)为基于电量方差特征示意图；(c)为基于IC曲线特征提取示意图；(d)为基于DV曲线特征提取示意图；(e)为基于DT曲线特征提取示意图；2 is a schematic diagram of a health factor of an embodiment; (a) is a schematic diagram of feature extraction based on measurement data; (b) is a schematic diagram based on power variance characteristics; (c) is a schematic diagram based on IC curve feature extraction; (d) is based on DV curve characteristics Extraction schematic diagram; (e) is a schematic diagram based on DT curve feature extraction;
图3为实施例技术路线图；3 is a technical roadmap of an embodiment;
图4为实施例采用的迁移深度学习网络结构图；Fig. 4 is the network structure diagram of migration deep learning adopted by the embodiment;
图5为实施例的健康因子筛选和模型建立流程。FIG. 5 is the health factor screening and model building process of the embodiment.
具体实施方式Detailed ways
以下通过特定的具体实例说明本发明的实施方式，本领域技术人员可由本说明书所揭露的内容轻易地了解本发明的其他优点与功效。本发明还可以通过另外不同的具体实施方式加以实施或应用，本说明书中的各项细节也可以基于不同观点与应用，在没有背离本发明的精神下进行各种修饰或改变。需要说明的是，以下实施例中所提供的图示仅以示意方式说明本发明的基本构想，在不冲突的情况下，以下实施例及实施例中的特征可以相互组合。The embodiments of the present invention are described below through specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments are only used to illustrate the basic idea of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.
其中，附图仅用于示例性说明，表示的仅是示意图，而非实物图，不能理解为对本发明的限制；为了更好地说明本发明的实施例，附图某些部件会有省略、放大或缩小，并不代表实际产品的尺寸；对本领域技术人员来说，附图中某些公知结构及其说明可能省略是可以理解的。Among them, the accompanying drawings are only used for exemplary description, and represent only schematic diagrams, not physical drawings, and should not be construed as limitations of the present invention; in order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings will be omitted, The enlargement or reduction does not represent the size of the actual product; it is understandable to those skilled in the art that some well-known structures and their descriptions in the accompanying drawings may be omitted.
本发明实施例的附图中相同或相似的标号对应相同或相似的部件；在本发明的描述中，需要理解的是，若有术语“上”、“下”、“左”、“右”、“前”、“后”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此附图中描述位置关系的用语仅用于示例性说明，不能理解为对本发明的限制，对于本领域的普通技术人员而言，可以根据具体情况理解上述术语的具体含义。The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms “upper”, “lower”, “left” and “right” , "front", "rear" and other indicated orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the indicated device or element must be It has a specific orientation, is constructed and operated in a specific orientation, so the terms describing the positional relationship in the accompanying drawings are only used for exemplary illustration, and should not be construed as a limitation of the present invention. situation to understand the specific meaning of the above terms.
请参阅图1，一种基于迁移深度学习的电池组剩余寿命预测方法可以分为以下步骤：Referring to Figure 1, a method for predicting the remaining life of a battery pack based on transfer deep learning can be divided into the following steps:
步骤S1：收集动力电池老化数据集，建立电池老化数据库，包括完整的电池单体全寿命周期数据以及由同款电芯组成的电池组早期的老化数据。Step S1: Collect power battery aging data sets, and establish a battery aging database, including complete battery life cycle data and early aging data of battery packs composed of cells of the same type.
步骤S2：根据电池单体老化数据提取多个健康因子，并根据相关性分析和容量估计误差筛选健康因子。Step S2: Extracting multiple health factors according to battery cell aging data, and screening health factors according to correlation analysis and capacity estimation error.
步骤S3：基于电池单体全寿命周期老化数据集训练得到健康因子的递推模型，以及基于健康因子的容量估计模型。Step S3: A recursive model of the health factor and a capacity estimation model based on the health factor are obtained by training based on the battery cell life cycle aging data set.
步骤S4：提取电池组早期老化实验数据集每个电池单体的健康因子，建立基于电池单体健康因子集和电池组容量衰减的机器学习模型。Step S4: extracting the health factor of each battery cell in the early aging experimental data set of the battery pack, and establishing a machine learning model based on the battery cell health factor set and the capacity decay of the battery pack.
步骤S5：利用迁移学习对电池组每个电池单体的健康因子递推模型进行微调，并进行健康因子外推预测，最后基于外推得到的单体健康因子集估计未来循环的电池组容量；基于单体容量估计模型预测未来各单体的容量，得到未来循环的电池组单体容量分布。Step S5: using transfer learning to fine-tune the recursive model of the health factor of each battery cell in the battery pack, and extrapolate the health factor to predict, and finally estimate the capacity of the battery pack in the future cycle based on the extrapolated cell health factor set; Based on the cell capacity estimation model, the capacity of each cell in the future is predicted, and the cell capacity distribution of the battery pack in the future cycle is obtained.
作为一种可选的实施例，本方案的完整技术路线图如图3所示。As an optional embodiment, the complete technical roadmap of this solution is shown in FIG. 3 .
作为一种可选的实施例，上述步骤S1具体包括S11-S13：As an optional embodiment, the above step S1 specifically includes S11-S13:
步骤S11：收集某款电芯的多个电池单体的全寿命周期完整数据集，包括充放电电流、电压、温度、时间、电量等参数，可涵盖不同充放电倍率和环境温度的数据集。Step S11 : Collect a complete data set of the life cycle of multiple battery cells of a certain type of cell, including parameters such as charge and discharge current, voltage, temperature, time, power, etc., which can cover data sets of different charge and discharge rates and ambient temperatures.
步骤S12：收集电池单体同款电芯对应的电池组的老化实验数据集，包括每个单体的电压、温度等参数和电池组电流、电量等信息。Step S12 : collecting an aging experiment data set of a battery pack corresponding to a cell of the same type of battery cell, including parameters such as voltage and temperature of each cell, and information such as battery pack current and power.
步骤S13：根据收集的电池单体和电池组的老化实验数据，建立某款电芯的电池老化数据集。Step S13 : establishing a battery aging data set of a certain type of battery cell according to the collected aging experimental data of the battery cell and the battery pack.
作为一种可选的实施例，上述步骤S2具体包括S21-S23：As an optional embodiment, the above step S2 specifically includes S21-S23:
步骤S21：根据电池单体全寿命周期提取多个可用的健康因子。Step S21: Extracting multiple available health factors according to the full life cycle of the battery cells.
步骤S22：通过相关性分析，筛选出和电池单体容量相关性高的健康因子。Step S22: Through correlation analysis, screen out health factors that are highly correlated with the battery cell capacity.
步骤S23：基于A号电池单体建立全寿命的健康因子估计容量的机器学习模型，并利用B号电池单体提取健康因子进行容量估计，分析估计误差。Step S23 : establishing a machine learning model for estimating the capacity by the health factor of the whole life based on the battery cell A, and using the battery cell B to extract the health factor to estimate the capacity, and analyze the estimation error.
作为一种可选的实施例，所述的S21中的多个可用的健康因子包括电压曲线斜率，等时间间隔电压/温度变化，等电压/温度间隔电量变化，电量序列方差，电量差方差，容量增量曲线峰值、谷值、峰间隔、峰面积，电压/温度差分曲线峰值、谷值、峰间隔等，如图2所示，为实施例的健康因子示意图；(a)为基于测量数据特征提取示意图；(b)为基于电量方差特征示意图；(c)为基于IC曲线特征提取示意图；(d)为基于DV曲线特征提取示意图；(e)为基于DT曲线特征提取示意图；As an optional embodiment, the multiple available health factors in S21 include the slope of the voltage curve, the voltage/temperature change at equal time intervals, the power change at equal voltage/temperature intervals, the power sequence variance, the power difference variance, The peak value, valley value, peak interval, peak area of the capacity increment curve, peak value, valley value, and peak interval of the voltage/temperature differential curve, as shown in Figure 2, are the schematic diagrams of the health factor of the embodiment; (a) is based on the measurement data Schematic diagram of feature extraction; (b) is a schematic diagram of feature extraction based on power variance; (c) is a schematic diagram of feature extraction based on IC curve; (d) is a schematic diagram of feature extraction based on DV curve; (e) is a schematic diagram of feature extraction based on DT curve;
提取方法流程如下：The extraction method flow is as follows:
电压曲线斜率；
b)相同充电/放电时间电压差：ETDV＝f(V0,t_interval)b) Voltage difference at the same charge/discharge time: ETDV=f(V 0 , t_interval)
c)相同电压区间电量差；EVDQ＝f(Q0,t_interval)c) Electricity difference in the same voltage interval; EVDQ=f(Q 0 , t_interval)
d)不同循环电量差方差；
e)增量容量(incremental capacity,IC)曲线特征(峰值、谷值、峰间距、峰面积等)；IC增量计算式：
f)差分电压(differential voltage,DV)曲线特征(谷值、峰值、峰间距等)；差分电压计算式：
g)差分温度(differential temperature,DT)曲线特征(谷值、峰值、峰间距等)。差分温度计算式：
作为一种可选的实施例，获取IC、DV、DT曲线过程中包括滤波降噪处理，采用高斯滤波：
作为一种可选的实施例，S22所述的评价体系中相关系数法可采用皮尔森相关系数，如下As an optional embodiment, the correlation coefficient method in the evaluation system described in S22 may adopt the Pearson correlation coefficient, as follows
式所示：
作为一种可选的实施例，采用高斯过程回归利用A号电池建模，利用B号电池评估估计As an optional embodiment, the Gaussian process regression is used to model the battery of size A, and the battery of size B is used to evaluate and estimate
效果，高斯过程回归算法流程如下：The effect of the Gaussian process regression algorithm is as follows:
假定输入输出符合以下贝叶斯多元回归模型：y＝f(x)+ε,
式中ε是符合高斯分布的白噪声。f(x)可写为：where ε is white noise with a Gaussian distribution. f(x) can be written as:
式中m(x)和k(x,x')分别为均值函数和协方差函数，分别为：where m(x) and k(x,x') are the mean function and covariance function, respectively, which are:
m(x)＝E[f(x)]，k(x,x')＝E[(f(x)-m(x))(f(x')-m(x'))T]m(x)=E[f(x)], k(x,x')=E[(f(x)-m(x))(f(x')-m(x')) T ]
输入输出关系式可写为：
作为一种可选的实施例，采用均方根误差评价估计效果。As an optional embodiment, the root mean square error is used to evaluate the estimation effect.
作为一种可选的实施例，上述步骤S3具体包括S31-S32As an optional embodiment, the foregoing step S3 specifically includes S31-S32
步骤S31：根据选定的健康因子，建立多个单体的全寿命周期健康因子递推衰减的深度学习模型。Step S31: According to the selected health factor, establish a deep learning model of recursive decay of the life cycle health factor of multiple monomers.
步骤S32：根据选的的健康因子，建立多个单体的全寿命周期基于健康因子的容量估计模型。Step S32: According to the selected health factor, establish a capacity estimation model based on the health factor for the whole life cycle of a plurality of monomers.
作为一种可选的实施例，全寿命周期健康因子递推衰减模型采用前m个健康因子预测下一循环的健康因子的方式，即：HIk+1＝f(HIk-m,...,HIk)，全寿命周期健康因子递推衰减模型利用深度神经网络进行建模，具体如图4所示，包括一层输入层，一层长短时记忆神经网络层，一层全连接层和一层输出层。容量估计模型同样采用高斯过程回归模型。As an optional embodiment, the whole life cycle health factor recursive decay model uses the first m health factors to predict the health factor of the next cycle, namely: HI k+1 =f(HI km ,..., HI k ), the whole life cycle health factor recursive decay model is modeled by using a deep neural network, as shown in Figure 4, including an input layer, a long and short-term memory neural network layer, a fully connected layer and a layer output layer. The capacity estimation model also adopts the Gaussian process regression model.
作为一种可选的实施例，步骤S4具体包括步骤S41-S42As an optional embodiment, step S4 specifically includes steps S41-S42
步骤S41：根据选定的健康因子，提取电池组早期循环中每个电池单体的健康因子，建立电池组各单体的健康因子组成的特征集。Step S41 : according to the selected health factor, extract the health factor of each battery cell in the early cycle of the battery pack, and establish a feature set composed of the health factor of each cell of the battery pack.
步骤S42：根据单体特征集和电池组容量真值，建立电池组容量估计的机器学习模型。Step S42: Establish a machine learning model for estimating battery pack capacity according to the cell feature set and the true value of the battery pack capacity.
作为一种可选的实施例，特征集为每个电池单体对应经筛选后的健康因子为一列，循环次数为行建立的特征矩阵，采用高斯过程回归建立基于特征集的电池组容量估计模型。As an optional embodiment, the feature set is a feature matrix in which the screened health factor corresponding to each battery cell is one column, and the number of cycles is a row. Gaussian process regression is used to establish a battery pack capacity estimation model based on the feature set. .
步骤S2-S4所述的具体技术实现如图5所示。The specific technical implementation described in steps S2-S4 is shown in FIG. 5 .
作为一种可选的实施例，步骤S5具体包括步骤S51-S54As an optional embodiment, step S5 specifically includes steps S51-S54
步骤S51：利用前期训练好的电池单体全寿命周期健康因子衰减模型和电池组各单体早期提取的健康因子，运用迁移学习的方法将训练好的网络进行部分结构再训练，得到电池组每个单体的健康因子递推模型。Step S51: Using the pre-trained battery cell life cycle health factor decay model and the health factors extracted early from each cell of the battery pack, use the transfer learning method to retrain the trained network for part of the structure, and obtain the battery pack each time. A recursive model of the health factor of a single individual.
步骤S52：根据每个单体的健康因子递推模型，外推得到未来未知循环次数的健康因子预测值。Step S52: According to the recursive model of the health factor of each monomer, extrapolate to obtain the predicted value of the health factor of the unknown number of cycles in the future.
步骤S53：根据建立的电池单体特征集估计电池组容量的机器学习模型，预测未来循环的电池组容量值。Step S53: According to the established battery cell feature set, a machine learning model for estimating the capacity of the battery pack is used to predict the capacity value of the battery pack in the future cycle.
步骤S54：根据建立的电池单体容量估计模型，得到未来未知循环的电池组内各单体的容量预测值，从而得到电池组单体的容量预测分布。Step S54 : according to the established battery cell capacity estimation model, obtain the capacity prediction value of each cell in the battery pack for unknown cycles in the future, so as to obtain the capacity prediction distribution of the battery pack cells.
作为一种可选的实施例，迁移学习冻结前面输入层和长短时记忆神经网络层，每个单体再训练时对全连接层和输出层的参数进行再训练，以达到模型微调的目的。As an optional embodiment, the transfer learning freezes the previous input layer and the long-term memory neural network layer, and retrains the parameters of the fully connected layer and the output layer when each unit is retrained, so as to achieve the purpose of fine-tuning the model.
最后说明的是，以上实施例仅用以说明本发明的技术方案而非限制，尽管参照较佳实施例对本发明进行了详细说明，本领域的普通技术人员应当理解，可以对本发明的技术方案进行修改或者等同替换，而不脱离本技术方案的宗旨和范围，其均应涵盖在本发明的权利要求范围当中。Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be Modifications or equivalent replacements, without departing from the spirit and scope of the technical solution, should all be included in the scope of the claims of the present invention.
Claims (4)
1.一种基于迁移深度学习的电池组剩余寿命预测方法，其特征在于：该方法包括以下步骤：1. A method for predicting the remaining life of a battery pack based on migration deep learning, characterized in that: the method comprises the following steps: 步骤S1：收集动力电池老化数据集，建立电池老化数据库，包括完整的电池单体全寿命周期数据以及由同款电芯组成的电池组早期的老化数据；Step S1: collect power battery aging data sets, and establish a battery aging database, including complete battery life cycle data and early aging data of battery packs composed of cells of the same type; 步骤S2：根据电池单体老化数据提取多个健康因子，并根据相关性分析和容量估计误差筛选健康因子；Step S2: extracting multiple health factors according to battery cell aging data, and screening health factors according to correlation analysis and capacity estimation error; 步骤S3：基于电池单体全寿命周期老化数据集训练得到健康因子的递推模型，以及基于健康因子的容量估计模型；Step S3: training a recursive model of the health factor and a capacity estimation model based on the health factor based on the battery cell life cycle aging data set; 步骤S4：提取电池组早期老化实验数据集每个电池单体的健康因子，建立基于电池单体健康因子集和电池组容量衰减的机器学习模型；Step S4: extracting the health factor of each battery cell in the battery pack early aging experiment data set, and establishing a machine learning model based on the battery cell health factor set and the battery pack capacity decay; 步骤S5：利用迁移学习对电池组每个电池单体的健康因子递推模型进行微调，并进行健康因子外推预测，最后基于外推得到的单体健康因子集估计未来循环的电池组容量；基于单体容量估计模型预测未来各单体的容量，得到未来循环的电池组单体容量分布；Step S5: using transfer learning to fine-tune the recursive model of the health factor of each battery cell in the battery pack, and extrapolate the health factor to predict, and finally estimate the capacity of the battery pack in the future cycle based on the extrapolated cell health factor set; Based on the cell capacity estimation model, the capacity of each cell in the future is predicted, and the cell capacity distribution of the battery pack in the future cycle is obtained; 所述步骤S1具体为：The step S1 is specifically: 步骤S11：收集某款电芯的多个电池单体的全寿命周期完整数据集，包括充放电电流、电压、温度、时间和电量参数，涵盖不同充放电倍率和环境温度的数据集；Step S11 : collect a complete data set of the full life cycle of a plurality of battery cells of a certain battery cell, including charge and discharge current, voltage, temperature, time and power parameters, and data sets covering different charge and discharge rates and ambient temperatures; 步骤S12：收集电池单体同款电芯对应的电池组的老化实验数据集，包括每个单体的电压和温度参数，以及电池组电流和电量信息；Step S12 : collecting the aging experiment data set of the battery pack corresponding to the same type of battery cell, including the voltage and temperature parameters of each cell, and the current and power information of the battery pack; 步骤S13：根据收集的电池单体和电池组的老化实验数据，建立某款电芯的电池老化数据集；Step S13 : establishing a battery aging data set of a certain type of battery cell according to the collected aging experimental data of the battery cell and the battery pack; 所述步骤S2具体为：The step S2 is specifically: 步骤S21：根据电池单体全寿命周期提取多个可用的健康因子；Step S21: extracting a plurality of available health factors according to the full life cycle of the battery cell; 步骤S22：通过相关性分析，筛选出和电池单体容量相关性高的健康因子；Step S22: Screening out health factors that are highly correlated with the battery cell capacity through correlation analysis; 步骤S23：基于A号电池单体建立全寿命的健康因子估计容量的机器学习模型，并利用B号电池单体提取健康因子进行容量估计，分析估计误差；Step S23 : establishing a machine learning model for estimating the capacity by the health factor of the whole life based on the battery cell A, and using the battery cell B to extract the health factor to estimate the capacity, and analyze the estimation error; 所述步骤S3具体为：The step S3 is specifically: 步骤S31：根据选定的健康因子，建立多个单体的全寿命周期健康因子递推衰减的深度学习模型；Step S31: According to the selected health factor, establish a deep learning model of recursive decay of the life cycle health factor of multiple monomers; 步骤S32：根据选的健康因子，建立多个单体的全寿命周期基于健康因子的容量估计模型；Step S32: According to the selected health factor, establish a capacity estimation model for the whole life cycle of a plurality of monomers based on the health factor; 所述步骤S4具体为：The step S4 is specifically: 步骤S41：根据选定的健康因子，提取电池组早期循环中每个电池单体的健康因子，建立电池组各单体的健康因子组成的特征集；Step S41: according to the selected health factor, extract the health factor of each battery cell in the early cycle of the battery pack, and establish a feature set composed of the health factor of each cell of the battery pack; 步骤S42：根据单体特征集和电池组容量真值，建立电池组容量估计的机器学习模型；Step S42: establishing a machine learning model for estimating battery pack capacity according to the cell feature set and the true value of the battery pack capacity; 所述步骤S5具体为：The step S5 is specifically: 步骤S51：利用前期训练好的电池单体全寿命周期健康因子衰减模型和电池组各单体早期提取的健康因子，运用迁移学习的方法将训练好的网络进行部分结构再训练，得到电池组每个单体的健康因子递推模型；Step S51: Using the pre-trained battery cell life cycle health factor decay model and the health factors extracted early from each cell of the battery pack, use the transfer learning method to retrain the trained network for part of the structure, and obtain the battery pack each time. A recursive model of health factors for a single individual; 步骤S52：根据每个单体的健康因子递推模型，外推得到未来未知循环次数的健康因子预测值；Step S52: According to the recursive model of the health factor of each monomer, extrapolate to obtain the predicted value of the health factor of the unknown number of cycles in the future; 步骤S53：根据建立的电池单体特征集估计电池组容量的机器学习模型，预测未来循环的电池组容量值；Step S53: according to the established battery cell feature set, a machine learning model for estimating the capacity of the battery pack, and predicting the capacity value of the battery pack in the future cycle; 步骤S54：根据建立的电池单体容量估计模型，得到未来未知循环的电池组内各单体的容量预测值，从而得到电池组单体的容量预测分布；Step S54: According to the established battery cell capacity estimation model, obtain the capacity prediction value of each cell in the battery pack for unknown cycles in the future, so as to obtain the capacity prediction distribution of the battery pack cells; 迁移学习的方法将基于早期训练好的健康因子递推模型的深度学习网络，选择性的冻结某些层，对剩下的层进行模型再训练，或拆掉某些层并重新构建网络进行新网络的再训练。The transfer learning method will be based on the deep learning network of the early trained health factor recursive model, selectively freeze some layers, retrain the model for the remaining layers, or remove some layers and rebuild the network for new Retraining of the network. 2.根据权利要求1所述的一种基于迁移深度学习的电池组剩余寿命预测方法，其特征在于：所述多个可用的健康因子包括电压曲线斜率，等时间间隔电压/温度变化，等电压/温度间隔电量变化，电量序列方差，电量差方差，容量增量曲线峰值、谷值、峰间隔、峰面积、电压/温度差分曲线峰值、谷值和峰间隔；2 . The method for predicting the remaining life of a battery pack based on migration deep learning according to claim 1 , wherein the plurality of available health factors include the slope of the voltage curve, voltage/temperature changes at equal time intervals, and equal voltages. 3 . / Temperature interval power change, power sequence variance, power difference variance, capacity increment curve peak, valley, peak interval, peak area, voltage/temperature difference curve peak, valley and peak interval; A号电池单体和B号电池单体指的是同款电芯的不同电池单体，机器学习模型为高斯过程回归模型或相关向量机模型，实现概率预测。A battery cell and a battery cell B refer to different battery cells of the same cell, and the machine learning model is a Gaussian process regression model or a correlation vector machine model to achieve probability prediction. 3.根据权利要求1所述的一种基于迁移深度学习的电池组剩余寿命预测方法，其特征在于：所述深度学习模型具体指多层神经网络构建的模型，为高斯过程回归模型或相关向量机模型，实现概率预测。3. The method for predicting the remaining life of a battery pack based on migration deep learning according to claim 1, wherein the deep learning model specifically refers to a model constructed by a multi-layer neural network, which is a Gaussian process regression model or a correlation vector machine model for probabilistic prediction. 4.根据权利要求1所述的一种基于迁移深度学习的电池组剩余寿命预测方法，其特征在于：所述特征集为电池组中每个电池单体特征组成的特征矩阵，为高斯过程回归模型或相关向量机模型，实现概率预测。4. The method for predicting the remaining life of a battery pack based on migration deep learning according to claim 1, wherein the feature set is a feature matrix composed of features of each battery cell in the battery pack, and is a Gaussian process regression Model or correlation vector machine model to achieve probabilistic prediction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110048627.4A CN112798960B (en) | 2021-01-14 | 2021-01-14 | A battery pack remaining life prediction method based on transfer deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110048627.4A CN112798960B (en) | 2021-01-14 | 2021-01-14 | A battery pack remaining life prediction method based on transfer deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112798960A CN112798960A (en) | 2021-05-14 |
| CN112798960B true CN112798960B (en) | 2022-06-24 |
Family
ID=75810843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110048627.4A Active CN112798960B (en) | 2021-01-14 | 2021-01-14 | A battery pack remaining life prediction method based on transfer deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112798960B (en) |
Families Citing this family (11)
* Cited by examiner, † Cited by third party| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113406496B (en) * | 2021-05-26 | 2023-02-28 | 广州市香港科大霍英东研究院 | Battery capacity prediction method, system, device and medium based on model migration |
| CN113567875A (en) * | 2021-06-11 | 2021-10-29 | 国电南瑞科技股份有限公司 | Ternary lithium ion battery health state estimation method based on K nearest neighbor regression |
| CN113820604B (en) * | 2021-08-30 | 2024-04-26 | 昆明理工大学 | Lithium battery SOH estimation method based on temperature prediction |
| CN113791351B (en) * | 2021-09-17 | 2022-04-19 | 电子科技大学 | Lithium battery life prediction method based on transfer learning and difference probability distribution |
| CN114167284B (en) * | 2021-11-02 | 2023-12-22 | 江苏博强新能源科技股份有限公司 | Lithium battery RUL prediction method and equipment based on BMS big data and integrated learning |
| CN114184972B (en) * | 2021-11-02 | 2023-12-22 | 江苏博强新能源科技股份有限公司 | Automatic estimation method and equipment for SOH of battery by combining data driving and electrochemical mechanism |
| CN114580262B (en) * | 2021-11-18 | 2024-07-12 | 吉林大学 | Lithium ion battery health state estimation method |
| CN114216558B (en) * | 2022-02-24 | 2022-06-14 | 西安因联信息科技有限公司 | Method and system for predicting remaining life of battery of wireless vibration sensor |
| WO2023189368A1 (en) * | 2022-03-30 | 2023-10-05 | ヌヴォトンテクノロジージャパン株式会社 | Storage battery degradation estimation device and storage battery degradation estimation method |
| CN115184829A (en) * | 2022-06-21 | 2022-10-14 | 华为数字能源技术有限公司 | A battery pack life prediction method and device |
| CN116593903B (en) * | 2023-07-17 | 2023-10-20 | 中国华能集团清洁能源技术研究院有限公司 | Battery remaining life prediction method and device |
Citations (11)
* Cited by examiner, † Cited by third party| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107064800A (en) * | 2016-11-29 | 2017-08-18 | 北京交通大学 | The real-time predicting method of lithium ion battery remaining life |
| CN107797067A (en) * | 2016-09-05 | 2018-03-13 | 北京航空航天大学 | Lithium ion battery life migration prediction method based on deep learning |
| CN109061504A (en) * | 2018-08-28 | 2018-12-21 | 中北大学 | Same type difference lithium ion battery remaining life prediction technique and system |
| CN109342949A (en) * | 2018-11-06 | 2019-02-15 | 长沙理工大学 | Online prediction method for remaining life of lithium-ion power battery during charging |
| CN110927591A (en) * | 2019-12-11 | 2020-03-27 | 北京理工大学 | A battery capacity estimation method, computer readable medium and vehicle |
| CN111007417A (en) * | 2019-12-06 | 2020-04-14 | 重庆大学 | Battery pack SOH and RUL prediction method and system based on inconsistency assessment |
| CN111443294A (en) * | 2020-04-10 | 2020-07-24 | 华东理工大学 | Method and device for indirect prediction of remaining life of lithium ion battery |
| CN111722115A (en) * | 2019-03-18 | 2020-09-29 | 上海汽车集团股份有限公司 | Power battery service life prediction method and system |
| CN111983474A (en) * | 2020-08-25 | 2020-11-24 | 陕西科技大学 | Lithium ion battery life prediction method and system based on capacity decline model |
| CN112014735A (en) * | 2019-05-30 | 2020-12-01 | 上海汽车集团股份有限公司 | Battery cell aging life prediction method and device based on full life cycle |
| CN112083337A (en) * | 2020-10-22 | 2020-12-15 | 重庆大学 | Power battery health prediction method oriented to predictive operation and maintenance |
Family Cites Families (4)
* Cited by examiner, † Cited by third party| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150349385A1 (en) * | 2014-04-01 | 2015-12-03 | Medtronic, Inc. | Method and System for Predicting Useful Life of a Rechargeable Battery |
| US11131713B2 (en) * | 2018-02-21 | 2021-09-28 | Nec Corporation | Deep learning approach for battery aging model |
| CN111638465B (en) * | 2020-05-29 | 2023-02-28 | 浙大宁波理工学院 | Lithium battery health state estimation method based on convolutional neural network and transfer learning |
| CN112051506B (en) * | 2020-08-28 | 2021-07-27 | 北京航空航天大学 | A similar product transferable sample screening method, system and use |
-
2021
- 2021-01-14 CN CN202110048627.4A patent/CN112798960B/en active Active
Patent Citations (11)
* Cited by examiner, † Cited by third party| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107797067A (en) * | 2016-09-05 | 2018-03-13 | 北京航空航天大学 | Lithium ion battery life migration prediction method based on deep learning |
| CN107064800A (en) * | 2016-11-29 | 2017-08-18 | 北京交通大学 | The real-time predicting method of lithium ion battery remaining life |
| CN109061504A (en) * | 2018-08-28 | 2018-12-21 | 中北大学 | Same type difference lithium ion battery remaining life prediction technique and system |
| CN109342949A (en) * | 2018-11-06 | 2019-02-15 | 长沙理工大学 | Online prediction method for remaining life of lithium-ion power battery during charging |
| CN111722115A (en) * | 2019-03-18 | 2020-09-29 | 上海汽车集团股份有限公司 | Power battery service life prediction method and system |
| CN112014735A (en) * | 2019-05-30 | 2020-12-01 | 上海汽车集团股份有限公司 | Battery cell aging life prediction method and device based on full life cycle |
| CN111007417A (en) * | 2019-12-06 | 2020-04-14 | 重庆大学 | Battery pack SOH and RUL prediction method and system based on inconsistency assessment |
| CN110927591A (en) * | 2019-12-11 | 2020-03-27 | 北京理工大学 | A battery capacity estimation method, computer readable medium and vehicle |
| CN111443294A (en) * | 2020-04-10 | 2020-07-24 | 华东理工大学 | Method and device for indirect prediction of remaining life of lithium ion battery |
| CN111983474A (en) * | 2020-08-25 | 2020-11-24 | 陕西科技大学 | Lithium ion battery life prediction method and system based on capacity decline model |
| CN112083337A (en) * | 2020-10-22 | 2020-12-15 | 重庆大学 | Power battery health prediction method oriented to predictive operation and maintenance |
Non-Patent Citations (1)
* Cited by examiner, † Cited by third party| Title |
|---|
| 锂离子电池组健康状态估计综述;刘大同 等;《仪器仪表学报》;20201231;第41卷(第11期);第1-18页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112798960A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112798960B (en) | 2022-06-24 | A battery pack remaining life prediction method based on transfer deep learning |
| CN111443294B (en) | 2022-09-23 | Method and device for indirectly predicting remaining life of lithium ion battery |
| CN110221225B (en) | 2021-02-26 | Spacecraft lithium ion battery cycle life prediction method |
| CN107957562B (en) | 2023-06-09 | An online prediction method for the remaining life of lithium-ion batteries |
| CN112083337B (en) | 2023-06-16 | A power battery health prediction method for predictive operation and maintenance |
| CN110222431B (en) | 2022-04-12 | Remaining life prediction method of lithium-ion battery based on fusion of gated recurrent unit neural network and Kalman filter model |
| CN110568359B (en) | 2021-11-23 | Lithium battery residual life prediction method |
| CN111007417A (en) | 2020-04-14 | Battery pack SOH and RUL prediction method and system based on inconsistency assessment |
| CN110175672A (en) | 2019-08-27 | The battery charging state assessment system and appraisal procedure combined based on Extended Kalman filter and Genetic BP Neutral Network |
| CN113219343A (en) | 2021-08-06 | Lithium battery health state prediction method, system, equipment and medium based on elastic network |
| CN112611976A (en) | 2021-04-06 | Power battery state of health estimation method based on double differential curves |
| CN113128672B (en) | 2022-11-29 | Lithium ion battery pack SOH estimation method based on transfer learning algorithm |
| CN113449919B (en) | 2022-12-09 | Power consumption prediction method and system based on feature and trend perception |
| CN112834927A (en) | 2021-05-25 | Method, system, device and medium for predicting remaining life of lithium battery |
| CN117825965A (en) | 2024-04-05 | State prediction method and system for lithium ion battery |
| CN114487890A (en) | 2022-05-13 | An improved long short-term memory neural network for lithium battery state of health estimation method |
| CN115236522A (en) | 2022-10-25 | End-to-end capacity estimation method of energy storage battery based on hybrid deep neural network |
| CN109633448A (en) | 2019-04-16 | Identify the method, apparatus and terminal device of cell health state |
| CN117648631B (en) | 2024-05-28 | Power battery health state estimation method for electric automobile group |
| CN113657030A (en) | 2021-11-16 | Method for predicting remaining service life of lithium battery based on Gaussian process regression |
| CN115453399A (en) | 2022-12-09 | A Battery Pack SOH Estimation Method Considering Inconsistency |
| CN115577637A (en) | 2023-01-06 | Knowledge and data fused power supply system residual life prediction method |
| CN118655466A (en) | 2024-09-17 | A lithium battery health state SOH prediction method, device and storage medium |
| CN117572250A (en) | 2024-02-20 | Method for estimating battery SOH based on multi-feature fusion and XGBoost |
| CN118169582A (en) | 2024-06-11 | Lithium ion battery health state and residual life prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 2021-05-14 | PB01 | Publication | |
| 2021-05-14 | PB01 | Publication | |
| 2021-06-01 | SE01 | Entry into force of request for substantive examination | |
| 2021-06-01 | SE01 | Entry into force of request for substantive examination | |
| 2022-06-24 | GR01 | Patent grant | |
| 2022-06-24 | GR01 | Patent grant | |
| 2024-12-27 | EE01 | Entry into force of recordation of patent licensing contract | |
| 2024-12-27 | EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210514 Assignee: Chongqing Delida New Energy Technology Co.,Ltd. Assignor: Chongqing University Contract record no.: X2024980031476 Denomination of invention: A method for predicting the remaining life of battery packs based on transfer deep learning Granted publication date: 20220624 License type: Common License Record date: 20241211 |
| 2024-12-31 | EE01 | Entry into force of recordation of patent licensing contract | |
| 2024-12-31 | EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210514 Assignee: Chongqing Jielan Industrial Co.,Ltd.|Chongqing Yeniu Measurement and Control Technology Co.,Ltd.|Chongqing Suren Intelligent Technology Co.,Ltd.|Zhaolong Intelligent Technology (Chongqing) Co.,Ltd.|Chongqing Benjing Technology Co.,Ltd.|Chongqing Siji Life Technology Co.,Ltd.|Chongqing Benjing Yuzhi Innovation Technology Co.,Ltd.|Chongqing Huhu Technology Co.,Ltd. Assignor: Chongqing University Contract record no.: X2024980036157 Denomination of invention: A method for predicting the remaining life of battery packs based on transfer deep learning Granted publication date: 20220624 License type: Common License Record date: 20241213 |