CN108303652A - A kind of lithium battery method for predicting residual useful life - Google Patents
A kind of lithium battery method for predicting residual useful life Download PDFInfo
- Publication number
- CN108303652A CN108303652A CN201810052667.4A CN201810052667A CN108303652A CN 108303652 A CN108303652 A CN 108303652A CN 201810052667 A CN201810052667 A CN 201810052667A CN 108303652 A CN108303652 A CN 108303652A
- Authority
- CN
- China
- Prior art keywords
- sequence
- battery
- prediction
- decomposition
- useful life
- 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
Links
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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses a kind of lithium battery method for predicting residual useful life, including two stages：First stage is catabolic phase, and complicated cell health state (State of Health, SOH) sequence is decomposed into limited a intrinsic mode functions (Intrinsic Mode Function, IMF) and a survival function using EMD.Second stage is forecast period, is predicted each function after decomposition using autoregression integral moving average model (ARIMA), and each prediction addition is obtained whole SOH predicted values again, and then obtains remaining battery life.The present invention fully considers that the localised waving part of battery SOH sequences, prediction are more authentic and valid.
Description
Technical field
The invention belongs to battery technology fields, are related to a kind of prediction technique of remaining battery life, and in particular to Yi Zhongrong
Close the lithium battery method for predicting residual useful life of empirical mode decomposition and autoregression integral moving average model.
Background technology
Electric vehicle is the inexorable trend of future automobile development.Due to the aging characteristics of power battery, continuous decrement continues
It is the restraining factors of a wide range of rapid proliferation of electric vehicle to sail mileage and safety problem always.Cell degradation not only causes battery to hold
The reduction of amount, has been greatly reduced the output of electric vehicle maximum instantaneous power, then causes automobile power insufficient, overtaking other vehicles and
There are security risks in the case of upward slope；The battery-heating of aging simultaneously is serious, and on fire caused by overheat is also electric vehicle peace
One of full hidden danger.Country《Energy saving and new-energy automobile Technology Roadmap》In explicitly point out the mid-term developing direction of power battery and be
Promote power battery safety, consistency and service life.The remaining life of battery refers generally to, and decays from current time to battery capacity
To manufacture capacity 80% when can continue the numbers of charge and discharge.Remaining life (the remaining useful of power battery
Life, RUL) prediction can effectively evade the safety accident caused by battery performance excessive attenuation, and can plan the dimension of battery in advance
It repaiies and replaces.
Currently, remaining battery life prediction technique is varied, there are modelling, data-driven method etc..Modelling Chang Yi electricity
Charge and discharge number in pond is that input obtains remaining battery life.Since the model in the method is usually by specific behaviour in service (such as temperature
Degree is certain, constant-current discharge) under obtain, therefore modelling limitation is larger.In recent years with the rise of machine learning, more and more
Data-driven method is applied to battery life predicting, such as neural network, support vector machines, although such method is compared with the former
Have the advantages that not depending on battery behaviour in service, but its precision of prediction is but closely related with battery data, also chooses with parameter
Related, although can more be calculated to a nicety value sometimes, prediction result is extremely unstable.
(capacity regeneration) is regenerated along with electricity in battery capacity attenuation process, this is remaining life
Predict the part that can not ignore.Electricity regenerates so that battery capacity decaying sequence is extremely unstable, therefore above two method is all
It is difficult to realize Accurate Prediction always.
Invention content
In order to solve the above technical problem, the present invention provides a kind of fusion empirical mode decompositions and autoregression integral sliding
The lithium battery method for predicting residual useful life of averaging model.
The technical solution adopted in the present invention is：A kind of lithium battery method for predicting residual useful life, it is characterised in that：Including dividing
Solution stage and forecast period；
The catabolic phase is that battery capacity sequence is decomposed into the different several miniature sequences of frequency and a dull sequence
Row, the former is intrinsic mode functions, and the latter is survival function；Miniature sequence represents battery capacity orthogenesis, and monotonic sequence generation
The attenuation trend of table battery capacity；
The forecast period includes following two steps：
Step 1：Individual forecast is carried out to each sequence after decomposition；
First determine whether sequence is stationary sequence, if unstable progress difference is until steady；Prediction mould is determined later
The exponent number and parameter of type, make the prediction of unique sequence；
Step 2：Integrated forecasting；
All Individual forecast sequences are added, the macro-forecast sequence of battery capacity is obtained；It will be at the beginning of the sequence and battery
80% comparison of beginning electricity, obtains the remaining life of battery.
Compared with the existing technology, the beneficial effects of the invention are as follows：
One, each sequence after decomposing saves former SOH sequence informations completely, and the sequence after decomposition not only contains cell decay
Overall trend also contains battery capacity and regenerates sequence；It can more reflect the actual conditions of cell decay based on this prediction.
Two, the sequence after decomposing is simpler than original SOH sequences, is easy to apply suitable prediction model, prediction more accurate.And
Conventional method directly gives a forecast to SOH sequences, since original series fluctuation is irregular, so prediction difficulty is big, low precision.
Description of the drawings
Fig. 1 is the flow chart of the embodiment of the present invention；
Fig. 2 is Individual forecast flow chart in the embodiment of the present invention；
Fig. 3 is battery SOH (electricity hundred-mark system) sequence after EMD is decomposed of the embodiment of the present invention.
Fig. 4 is the SOH sequence prediction results of the embodiment of the present invention.
Specific implementation mode
Understand for the ease of those of ordinary skill in the art and implement the present invention, with reference to the accompanying drawings and embodiments to this hair
It is bright to be described in further detail, it should be understood that implementation example described herein is merely to illustrate and explain the present invention, not
For limiting the present invention.
Battery capacity sequence can be decomposed into the different several miniature sequences of frequency and a dull sequence by empirical mode decomposition
Row, the former is intrinsic mode functions, and the latter is survival function.Miniature sequence represents battery capacity orthogenesis, and monotonic sequence
Represent the attenuation trend of battery capacity.Each sequence is individually predicted after decomposition with ARIMA, the electricity regeneration for both having considered battery is existing
As, and with reference to cell decay trend.Therefore theoretically, this kind has merged empirical mode decomposition and autoregression integral sliding average
The lithium battery method for predicting residual useful life of model can significantly improve estimated performance.
Based on above-mentioned theory, the present invention, which proposes, a kind of having merged empirical mode decomposition (empirical mode
Decomposition, EMD) and autoregression integral moving average model (autoregressive integrated moving
Average, ARIMA) lithium battery method for predicting residual useful life.
Referring to Fig.1, the present invention includes two stages：First stage is catabolic phase, and application experience mode decomposition (EMD) will
Complicated cell health state (State-of-Health, SOH) sequence is decomposed into limited a intrinsic mode functions (Intrinsic
Mode Function, IMF) and a survival function.Second stage is forecast period, and sliding average is integrated using autoregression
Model (ARIMA) carries out Individual forecast to each function after decomposition, and each prediction addition is obtained whole SOH again and is predicted
Value, and then obtain remaining battery life.
For ARIMA models, see Fig. 2, Individual forecast of the invention specific implementation process is：
Step 1.1：Judge whether sequence is stationary sequence；
If so, thening follow the steps 1.2；
If it is not, then carrying out difference to sequence until steadily, then executing step 1.2；
Step 1.2：Determine the exponent number of ARIMA models；
Step 1.3：The parameter of ARIMA models is determined using least square method；
Step 1.4：Make the prediction of unique sequence；
Step 1.5：Judge whether to reach end of life；
If so, this flow terminates；
If it is not, future position is then added to training sequence, and turns round and execute step 1.3.
The present embodiment is by battery capacity divided by initial quantity of electricity, and hundred-mark system obtains battery SOH sequences.Using EMD to battery
SOH sequences are decomposed, and each sequence shown in Fig. 3 is obtained.ARIMA is used to be tied to each sequence prediction and by each prediction later
Fruit is added to obtain prediction result as shown in Figure 4.Then, predicted value is made comparisons with failure threshold (the 80% of initial quantity of electricity),
It is remaining life to be originated from prediction and reach the charge and discharge number of failure threshold.
It should be noted that the battery capacity sequence decomposition of the present embodiment is not limited to empirical mode decomposition, other decomposition
Method such as wavelet decomposition, set empirical mode decomposition are also by rights protection.
It should be noted that the unique sequence prediction technique of the present embodiment, is not limited solely to ARIMA predictions, can also make
It is predicted with other methods and RVM, SVM, Gaussian process regression model.
It should be understood that the part that this specification does not elaborate belongs to the prior art.
It should be understood that the above-mentioned description for preferred embodiment is more detailed, can not therefore be considered to this
The limitation of invention patent protection range, those skilled in the art under the inspiration of the present invention, are not departing from power of the present invention
Profit requires under protected ambit, can also make replacement or deformation, each fall within protection scope of the present invention, this hair
It is bright range is claimed to be determined by the appended claims.
Claims (4)
1. a kind of lithium battery method for predicting residual useful life, it is characterised in that：Including catabolic phase and forecast period；
The catabolic phase is that battery capacity sequence is decomposed into the different several miniature sequences of frequency and a monotonic sequence,
The former is intrinsic mode functions, and the latter is survival function；Intrinsic mode functions represent battery capacity orthogenesis, and survival function represents
The attenuation trend of battery capacity；
The forecast period includes following two steps：
Step 1：Individual forecast is carried out to each sequence after decomposition；
First determine whether sequence is stationary sequence, if unstable progress difference is until steady；Prediction model is determined later
Exponent number and parameter make the prediction of unique sequence；
Step 2：Integrated forecasting；
All Individual forecast sequences are added, the macro-forecast sequence of battery capacity is obtained；The sequence and battery is initially electric
80% comparison of amount, obtains the remaining life of battery.
2. lithium battery method for predicting residual useful life according to claim 1, it is characterised in that：The catabolic phase, will be electric
Pond electricity sequence decomposition method includes empirical mode decomposition EMD, wavelet decomposition, set empirical mode decomposition.
3. lithium battery method for predicting residual useful life according to claim 1, it is characterised in that：Mould is predicted described in step 1
Type includes ARIMA models, RVM models, SVM models, Gaussian process regression model.
4. lithium battery method for predicting residual useful life according to claim 3, which is characterized in that the Individual forecast, for
ARIMA prediction models, specific implementation includes following sub-step：
Step 1.1：Judge whether sequence is stationary sequence；
If so, thening follow the steps 1.2；
If it is not, then carrying out difference to sequence until steadily, then executing step 1.2；
Step 1.2：Determine the exponent number of prediction model；
Step 1.3：The parameter of prediction model is determined using least square method；
Step 1.4：Make the prediction of unique sequence；
Step 1.5：Judge whether to reach end of life；
If so, this flow terminates；
If it is not, future position is then added to training sequence, and turns round and execute step 1.3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810052667.4A CN108303652A (en) | 2018-01-18 | 2018-01-18 | A kind of lithium battery method for predicting residual useful life |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810052667.4A CN108303652A (en) | 2018-01-18 | 2018-01-18 | A kind of lithium battery method for predicting residual useful life |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108303652A true CN108303652A (en) | 2018-07-20 |
Family
ID=62865921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810052667.4A Pending CN108303652A (en) | 2018-01-18 | 2018-01-18 | A kind of lithium battery method for predicting residual useful life |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108303652A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109031153A (en) * | 2018-10-16 | 2018-12-18 | 北京交通大学 | A kind of health status On-line Estimation method of lithium ion battery |
| CN110109028A (en) * | 2019-04-12 | 2019-08-09 | 江苏大学 | A kind of power battery remaining life indirect predictions method |
| CN110221225A (en) * | 2019-07-08 | 2019-09-10 | 中国人民解放军国防科技大学 | Spacecraft lithium ion battery cycle life prediction method |
| CN110555226A (en) * | 2019-04-03 | 2019-12-10 | 太原理工大学 | method for predicting residual life of lithium iron phosphate battery based on EMD and MLP |
| CN110888077A (en) * | 2019-10-30 | 2020-03-17 | 无锡市产品质量监督检验院 | Accelerated lithium ion battery life evaluation method based on ARIMA time sequence |
| CN111143973A (en) * | 2019-12-05 | 2020-05-12 | 云南电网有限责任公司玉溪供电局 | Valve-regulated lead-acid storage battery degradation trend prediction method based on Gauss process regression |
| CN111157897A (en) * | 2019-12-31 | 2020-05-15 | 国网北京市电力公司 | Method and device for evaluating power battery, storage medium and processor |
| CN112415414A (en) * | 2020-10-09 | 2021-02-26 | 杭州电子科技大学 | Method for predicting remaining service life of lithium ion battery |
| CN112540317A (en) * | 2020-12-16 | 2021-03-23 | 武汉理工大学 | Battery health state estimation and residual life prediction method based on real vehicle data |
| CN112666483A (en) * | 2020-12-29 | 2021-04-16 | 长沙理工大学 | Improved ARMA lithium battery residual life prediction method |
| CN113406525A (en) * | 2021-06-15 | 2021-09-17 | 安庆师范大学 | Lithium battery pack residual life prediction method based on optimized variational modal decomposition |
| CN113657012A (en) * | 2021-07-21 | 2021-11-16 | 西安理工大学 | TCN and particle filter-based method for predicting residual life of key equipment |
| CN115032541A (en) * | 2022-04-13 | 2022-09-09 | 北京理工大学 | Lithium ion battery capacity degradation prediction method and device |
| CN115808627A (en) * | 2023-02-03 | 2023-03-17 | 泉州装备制造研究所 | Lithium battery SOH prediction method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104459560A (en) * | 2014-12-29 | 2015-03-25 | 合肥工业大学 | Method for predicting remaining service life of lithium battery based on wavelet denoising and relevant vector machine |
| CN105404939A (en) * | 2015-12-04 | 2016-03-16 | 河南许继仪表有限公司 | Short-term power load prediction method |
| CN107015165A (en) * | 2017-06-14 | 2017-08-04 | 安庆师范大学 | Lithium battery method for predicting residual useful life based on sparse coefficient multinuclear Method Using Relevance Vector Machine |
| CN107464226A (en) * | 2017-07-31 | 2017-12-12 | 东南大学 | A kind of image de-noising method based on improvement two-dimensional empirical mode decomposition algorithm |
-
2018
- 2018-01-18 CN CN201810052667.4A patent/CN108303652A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104459560A (en) * | 2014-12-29 | 2015-03-25 | 合肥工业大学 | Method for predicting remaining service life of lithium battery based on wavelet denoising and relevant vector machine |
| CN105404939A (en) * | 2015-12-04 | 2016-03-16 | 河南许继仪表有限公司 | Short-term power load prediction method |
| CN107015165A (en) * | 2017-06-14 | 2017-08-04 | 安庆师范大学 | Lithium battery method for predicting residual useful life based on sparse coefficient multinuclear Method Using Relevance Vector Machine |
| CN107464226A (en) * | 2017-07-31 | 2017-12-12 | 东南大学 | A kind of image de-noising method based on improvement two-dimensional empirical mode decomposition algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 岳相臣: "《经验模态分解算法应用研究》", 《硕士学位论文》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109031153B (en) * | 2018-10-16 | 2020-01-24 | 北京交通大学 | Online health state estimation method for lithium ion battery |
| CN109031153A (en) * | 2018-10-16 | 2018-12-18 | 北京交通大学 | A kind of health status On-line Estimation method of lithium ion battery |
| CN110555226A (en) * | 2019-04-03 | 2019-12-10 | 太原理工大学 | method for predicting residual life of lithium iron phosphate battery based on EMD and MLP |
| CN110109028A (en) * | 2019-04-12 | 2019-08-09 | 江苏大学 | A kind of power battery remaining life indirect predictions method |
| CN110221225B (en) * | 2019-07-08 | 2021-02-26 | 中国人民解放军国防科技大学 | Spacecraft lithium ion battery cycle life prediction method |
| CN110221225A (en) * | 2019-07-08 | 2019-09-10 | 中国人民解放军国防科技大学 | Spacecraft lithium ion battery cycle life prediction method |
| CN110888077A (en) * | 2019-10-30 | 2020-03-17 | 无锡市产品质量监督检验院 | Accelerated lithium ion battery life evaluation method based on ARIMA time sequence |
| CN111143973B (en) * | 2019-12-05 | 2021-01-26 | 云南电网有限责任公司玉溪供电局 | Valve-regulated lead-acid storage battery degradation trend prediction method based on Gauss process regression |
| CN111143973A (en) * | 2019-12-05 | 2020-05-12 | 云南电网有限责任公司玉溪供电局 | Valve-regulated lead-acid storage battery degradation trend prediction method based on Gauss process regression |
| CN111157897B (en) * | 2019-12-31 | 2022-05-10 | 国网北京市电力公司 | Method and device for evaluating power battery, storage medium and processor |
| CN111157897A (en) * | 2019-12-31 | 2020-05-15 | 国网北京市电力公司 | Method and device for evaluating power battery, storage medium and processor |
| CN112415414A (en) * | 2020-10-09 | 2021-02-26 | 杭州电子科技大学 | Method for predicting remaining service life of lithium ion battery |
| CN112540317A (en) * | 2020-12-16 | 2021-03-23 | 武汉理工大学 | Battery health state estimation and residual life prediction method based on real vehicle data |
| CN112540317B (en) * | 2020-12-16 | 2022-12-02 | 武汉理工大学 | Battery health state estimation and residual life prediction method based on real vehicle data |
| CN112666483B (en) * | 2020-12-29 | 2022-06-21 | 长沙理工大学 | Lithium battery residual life prediction method for improving ARMA (autoregressive moving average) |
| CN112666483A (en) * | 2020-12-29 | 2021-04-16 | 长沙理工大学 | Improved ARMA lithium battery residual life prediction method |
| CN113406525A (en) * | 2021-06-15 | 2021-09-17 | 安庆师范大学 | Lithium battery pack residual life prediction method based on optimized variational modal decomposition |
| CN113406525B (en) * | 2021-06-15 | 2023-06-02 | 金陵科技学院 | Lithium battery pack residual life prediction method based on optimization variation modal decomposition |
| CN113657012A (en) * | 2021-07-21 | 2021-11-16 | 西安理工大学 | TCN and particle filter-based method for predicting residual life of key equipment |
| CN115032541A (en) * | 2022-04-13 | 2022-09-09 | 北京理工大学 | Lithium ion battery capacity degradation prediction method and device |
| CN115032541B (en) * | 2022-04-13 | 2023-12-05 | 北京理工大学 | Lithium ion battery capacity degradation prediction method and device |
| CN115808627A (en) * | 2023-02-03 | 2023-03-17 | 泉州装备制造研究所 | Lithium battery SOH prediction method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108303652A (en) | A kind of lithium battery method for predicting residual useful life | |
| Chen et al. | A novel energy management method for series plug-in hybrid electric vehicles | |
| Zhu et al. | Adaptive energy management of a battery-supercapacitor energy storage system for electric vehicles based on flexible perception and neural network fitting | |
| CN108110349B (en) | Battery charging method and device and computer readable storage medium | |
| CN115291116B (en) | Energy storage battery health state prediction method and device and intelligent terminal | |
| CN102044717A (en) | Maximum charge-discharge power control method of lithium-ion battery pack | |
| CN104408257A (en) | Simulated annealing particle swarm algorithm based hybrid power automobile parameter optimization method | |
| CN110364775A (en) | Charge control method, battery charge controller and the terminal of battery | |
| CN115303290B (en) | System key level switching method and system of vehicle hybrid key level system | |
| Wei et al. | State of health assessment for echelon utilization batteries based on deep neural network learning with error correction | |
| CN114740388A (en) | Lithium battery residual life state evaluation method based on improved TCN | |
| CN110648013A (en) | Electric vehicle charging load prediction method based on dual-mode maximum entropy | |
| CN117207905A (en) | Power supply operation and maintenance method, system and storage medium of new energy automobile | |
| Zhang et al. | A comparison study on electric vehicle growth forecasting based on grey system theory and NAR neural network | |
| CN106060668A (en) | Smart television setting method and device | |
| CN109818089A (en) | A kind of battery current-limiting method and device | |
| CN115499315A (en) | OTA (over the air) upgrading task execution method and device and electronic equipment | |
| CN113725863A (en) | Power grid autonomous control and decision method and system based on artificial intelligence | |
| CN116298944A (en) | Charging remaining time display control method, device, equipment, vehicle and medium | |
| Denathayalan et al. | AMRIT-Alternative Machine Reasoning and Integrative Techniques for prognostics model of electric vehicles | |
| Donadee et al. | Estimating the rate of battery degradation under a stationary Markov operating policy | |
| CN111071102A (en) | Flexible charging method and device of direct current charging pile | |
| CN116579902B (en) | Digital park electric carbon data mapping method, system, equipment and storage medium | |
| Xu et al. | Online health estimate of hybrid energy storage system based on fuzzy brain emotional learning neural networks | |
| CN117674096A (en) | Electric automobile charging load prediction method and terminal |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180720 |