A BAYESIAN APPROACH TO BATTERY PROGNOSTICS AND HEALTH MANAGEMENT

This chapter deals with each of three components of battery health management (BHM): battery state estimation, battery prognostics, and decision making. The state‐of‐health (SOH) of batteries is a measurement that reflects the general condition of a battery and its ability to deliver the specified performance compared to a fresh battery. The chapter discusses how to determine the remaining useful life (RUL) of a battery. State estimation techniques, like the extended Kalman filter (EKF), have been applied for real‐time prediction of state‐of‐charge (SOC) and SOH of automotive batteries. A decision‐level fusion of data‐driven algorithms, like Autoregressive Integrated Moving Average (ARIMA) and neural networks, have been investigated for both diagnostics and prognostics. Particle filters (PFs) are a novel class of nonlinear filters that combine Bayesian learning techniques with importance sampling to provide good state tracking performance while keeping the computational load tractable.