A Rao-Blackwellised Unscented Kalman Filtering for MPPT Estimation in Photovoltaic Systems

Maximum power point tracking (MPPT) technology suits photovoltaic (PV) systems analysis well. It has better improvement for power generation efficiency and control effectiveness. However, the inherent nonlinear characteristics of the photovoltaic system and external factors such as irradiance and temperature hinder the stable system operation at the maximum power point (MPP). In addition, the accuracy and complexity of the PV system modeling are crucial to the performance of the PV system. This paper develops a Rao-Blackwellised Unscented Kalman Filtering (RBUKF) method for MPPT. Specifically, We first apply the Lambert W function to represent the current as an explicit function of its voltage, avoiding the need for the iterative solution and thus achieving faster and more accurate execution. Then, to improve the performance of the Unscented Kalman Filter (UKF), the Rao-Blackwellised method is applied to dynamic systems with nonlinear equations of state and linear equations of measurement. The simulation results show that the proposed method outperforms the traditional UKF in different transient and steady-state cases.