A Data-Driven Online SOP Estimation Method for Lithium-ion Capacitors

Lithium-ion capacitors (LICs) show promising applications in renewable energy integration and transportation electrification, due to their excellent performance in terms of energy density, power density and cycle life. Modeling and state of power (SOP) estimation are the keys to exploit advantages of LICs. A data-driven online SOP estimation method for lithiumion capacitors is proposed in this paper, the input vector of neural network is determined according to correlation analysis, and the optimal number of hidden layer nodes is selected by empirical formula. The simulation results show that the average absolute error of SOP estimation under Federal Urban Driving Schedule (FUDS) is less than 0.13W, and the root mean square error is less than 0.17W.