A Novel SOH Prediction Framework for the Lithium-ion Battery Using Echo State Network

Li-ion batteries provide lightweight, high energy density power sources for a variety of devices. Therefore, monitoring battery health in an effective way could increase the reliability and stability of the prediction system. So in this paper, we present a novel prediction framework based on Echo State Network to realize the prediction for battery state of health by training and testing battery impedance values and capacity values. To evaluate the proposed prediction approach, we have executed experiments with lithium-ion battery. Experimental results prove its effectiveness and confirm the estimation system can be effectively applied to the battery health state prediction. Moreover, the prediction system can run multiple data sets at a time to make the estimation process more efficient. Therefore, we can choose a battery which meets the requirement through the comparison between different batteries’ prediction results.