A Multirate-Fused State-of-Charge Estimation Scheme of Lithium-Ion Batteries for Electric Vehicles With Energy Harvesting Sensors

Abstract

State-of-charge (SOC), as a fundamental indicator for the remaining capacity of lithium-ion batteries (LBs) in electric vehicles, is of great significance for LBs’ operation optimization and life extension. In this article, the issue of multirate fusion estimation of SOC is addressed for LBs with energy harvesting sensors. To design the fused multirate estimation scheme of SOC, the adopted equivalent circuit model consisting of the resistor-capacitor networks, battery current and voltage, and ohmic resistance is first constructed. Then, the missing rate of the energy harvesting sensor and the probability distribution of the energy level carried by the sensor are recursively computed. To facilitate the estimator design, the multirate system is transformed into a single-rate system by virtue of the lifting technique. The objective of this article is to develop a set of local SOC estimators where the upper bound on each local estimation error covariance (EEC) is first guaranteed and subsequently is minimized by appropriately parameterizing the estimator gain. Next, the derived local estimates of SOC are fused using the covariance intersection (CI) fusion scheme. Finally, the proposed fused multirate SOC estimation scheme is illustrated by the simulation under federal urban driving schedule (FUDS) and dynamic stress test (DST) driving cycle.