Title:
LITHIUM BATTERY CAPACITY ESTIMATION METHOD BASED ON IMPROVED CONVOLUTION-LONG SHORT TERM MEMORY NEURAL NETWORK
Document Type and Number:
WIPO Patent Application WO/2021/138925
Kind Code:
A1
Abstract:
A lithium battery capacity estimation method based on an improved convolution-long short term memory neural network, relating to the technical field of lithium batteries. A lithium battery capacity estimation model is obtained by means of four steps, i.e., processing lithium battery data, adjusting hyper-parameters of an improved convolutional-long short term memory (CNN-LSTM) neural network by means of a genetic algorithm, training the improved CNN-LSTM neural network, and testing a model. An empirical mode decomposition algorithm is introduced to decompose the lithium battery data, so as to implement data de-noising. The hyper-parameters of the improved CNN-LSTM neural network are optimized by the genetic algorithm. Spatial features of charge and discharge data of the lithium battery are extracted using a convolutional neural network, then these features are inputted into the improved long short term memory neural network to extract temporal features, and finally an estimated capacity is outputted by means of a fully connected layer. The limitation that a conventional model-based algorithm excessively depends on a battery model is overcome; moreover, the prediction precision is high, and certain engineering applicability is achieved.
Inventors:
LI PENGHUA (CN)
ZHAGN ZIJIAN (CN)
WANG PING (CN)
XIONG QINGYU (CN)
SHAO ZIXUAN (CN)
HOU JIE (CN)
CHENG JIAWEI (CN)
ZHAGN ZIJIAN (CN)
WANG PING (CN)
XIONG QINGYU (CN)
SHAO ZIXUAN (CN)
HOU JIE (CN)
CHENG JIAWEI (CN)
Application Number:
PCT/CN2020/072069
Publication Date:
July 15, 2021
Filing Date:
January 14, 2020
Export Citation:
Assignee:
UNIV CHONGQING POSTS & TELECOM (CN)
International Classes:
G01R31/388; G01R31/367; G01R31/392; G01R31/396
Foreign References:
| CN109001640A | 2018-12-14 | |||
| CN109143105A | 2019-01-04 | |||
| CN109459699A | 2019-03-12 | |||
| CN110542866A | 2019-12-06 | |||
| CN110579710A | 2019-12-17 | |||
| US20160041229A1 | 2016-02-11 |
Attorney, Agent or Firm:
TIDYTEND INTELLECTUAL PROPERTY LAW FIRM (CN)
Download PDF: