A Battery Pack SOH Estimation Method Based on LSTM Neural Network

Abstract The invention provides a battery pack SOH estimation method based on LSTM neural network, which comprises the following steps. Firstly, collecting battery pack charging data of an electric vehicle which has actually operated for more than one year, performing correlation analysis, and extra...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wei, Tao, Liang, Jun, Chen, Weihe, Li, Yaotai, Shen, Xiaoyu, Cai, Tao, Wang, Limei, Xue, Anrong, Pan, Chaofeng, He, Zhigang, Tao, Yuanxue
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Abstract The invention provides a battery pack SOH estimation method based on LSTM neural network, which comprises the following steps. Firstly, collecting battery pack charging data of an electric vehicle which has actually operated for more than one year, performing correlation analysis, and extracting input characteristics of charging data, which are then used as the input of LSTM neural network. Then, SOC- electrical energy gain method is used to calculate the maximum available capacity of battery pack as the output feature of LSTM neural network. An LSTM neural network model is established and then that output value of the LSTM neural network is determined. Finally, the trained and verified LSTM neural network model is used to estimate SOH of battery pack. In this invention, the input and output characteristics of the LSTM neural network are obtained from the charging data of the battery pack of the electric vehicle which has been running for more than one year. This method is suitable for the SOH estimation of the battery pack when the electric vehicle is running in all state and all climate and overcomes the problem that the estimation method based on laboratory verification is difficult to adapt to the complex actual working environment. ht Ct-1 +Ct I t~nh I I ftJ0 x x CY rah CT ht_1 h xt Figure 1 A schematic diagram of the LSTM neural network model according to the present invention.