Neural network-based feedforward maximum power point tracking control for IDB converter-supplied PV system

Abstract

The photovoltaic generator exhibits a nonlinear v - i characteristic and its maximum power point varies with solar insolation. An intermediate converter is required to match the photovoltaic generator to the load and to operate the solar cell array at maximum power point. In this paper, a feedforward current sensorless maximum power point tracking scheme is developed for the interleaved dual boost converter-supplied photovoltaic system using only solar cell array information. For a given solar insolation, the tracking algorithm adjusts the solar cell array voltage to a value corresponding to the maximum power point. This is done by the feedforward loop, by comparing the instantaneous array voltage and a reference voltage corresponding to the maximum power point. The reference voltage for the feedforward loop is obtained by an off-line trained neural network. Experimental data is used for off-line training of the neural network, employing a back-propagation algorithm. A PI controller generates the control signal in real time for the PWM modulator and adjusts the duty ratio of the converter. The proposed current sensorless peak power tracking effectiveness is demonstrated through simulation and experimental results.