A Hybrid Data-Driven Approach for Multistep Ahead Prediction of State of Health and Remaining Useful Life of Lithium-Ion Batteries

Table 4

RUL prediction accuracy (%) for B0006.

RUL; Parameters	Cycle numbers
	50	55	60	65	70	75	80	85	90	95	100	105	110	115	120
SOH	97.64	98.43	89.76	84.25	81.1	81.89	85.83	84.25	84.25	94.49	92.13	97.64	93.7	95.28	96.06
IMF 1	100	48.82	70.87	75.59	81.89	81.1	86.61	93.7	73.23	66.14	74.02	79.53	90.55	80.31	94.49
IMF 2	93.7	99.21	40.16	46.46	48.82	68.5	94.49	11.81	98.43	33.07	98.43	14.17	22.05	18.9	32.28
IMF 3	10.24	20.47	8.66	18.9	40.16	37.8	14.96	30.71	72.44	77.17	59.84	52.76	65.35	85.04	99.21
Residual	96.06	93.7	91.34	89.76	88.98	88.19	87.4	88.19	88.98	89.76	91.34	92.91	94.49	96.85	98.43