An all digital speed adaptive maximum power point tracker for automotive photovoltaic applications

This paper proposes an all digital speed and step size adaptive maximum power point tracker (MPPT) for photovoltaic (PV) power generation, supporting the power supply of vehicles. To harvest significant amounts of energy in this application the MPPT has to deal with fast changing irradiance and severe shading. It has to achieve high tracking speed while maintaining tracking errors below 1%. The presented modified perturb and observe (P&O) MPPT algorithm meets these requirements by applying digital filtering to detect the settling of the input voltage automatically. This enables minimal sampling time for each situation and, hence, very fast convergence to the maximum power point (MPP). The developed algorithm has been implemented on a field-programmable gate array (FPGA) and verified with an experimental setup. It achieves about 9 kHz average switching frequency while maintaining a tracking accuracy above 99.8%.