Kalman Filter Based State of Charge Estimation for Valve Regulated Lead Acid Batteries in Wind Power Smoothing Applications

The anticipated increase in electrical power generation from wind and other renewable energy sources is expected to require new techniques to meet challenges faced as a result of the variability of these sources. Included in these techniques is the possibility of wind smoothing using battery energy storage systems. This thesis addresses the issue of state of charge estimation for such applications.;Traditional techniques used for state of charge (SOC) estimation are not well suited to this application due to the fact that the battery frequently fluctuates between charge and discharge. Recently some success has been reported using Kalman filter based SOC estimation in hybrid electric vehicle applications, an application that suffer from the same drawback.;This thesis also attempts to improve upon the state of the art in Kalman filter based SOC estimation by developing a new model of the valve regulated lead acid (VRLA) battery. The model describes the time varying voltage current relationship of the battery using a set of ordinary differential equations (ODEs), and is used as the basis for a new Kalman filter based SOC estimator. The model is derived using simplifying assumptions on a more complex electrochemistry based battery model.;The thesis also describes the inclusion of the model within a Kalman filter based state of charge (SOC) estimator. The new lumped electrochemical (LEC) model and the related estimator are then compared to the state of the art using two profiles designed to emulate the behavior of a battery within wind smoothing applications. The new estimator based on the LEC model shows clear improvement in performance over the state of the art.