Analysis of State-of-Health Estimation Approaches and Constraints for Lithium-Ion Batteries in Electric Vehicles

Table 4

Average error of different SOH methods.

Model	Error rate (in %)	Error type	Reference
Elman NN	1.29	MAE	[133]
Semisupervised transfer component analysis	1.29	MAE	[134]
ICA	2.99	RMSE	[135]
RF	3.58	RMSE	[136]
SVM	3.62	RMSE	[137]
SVM	0.13 and 024	MAE and RMSE	[138]
CNN	1 and 2	MAE and RMSE	[139]
RNN	0.72-1.37	RMSE	[140]
ELM	2	RMSE	[141]
ELM	1.93	MAE	[142]
ELM	0.4	MAE	[143]
ELM	1	MAE	[144]
LSTM	0.23	RMSE	[145]
LSTM	4.26	MAE	[146]
LSTM	2.16	Average RMSE	[147]
LSTM	2	MAE	[148]