DESIGN AND SIMULATION OF IMPROVED PI NEURON CONTROL FOR PHOTOVOLTAIC INVERTER

Zhou Rui, Zhang Yu, Zhang Lie-ping

ABSTRACT: The algorithm of traditional double closed loop PI control has the advantages of simple structure, good robustness and high reliability, but its parameter setting is difficult, and not good at tracking changes of photovoltaic power generation system in complex nonlinear case. Aiming at the above problems, a control algorithm based on neuron PI inverter has been proposed. Neurons with proportional and integral functions are defined, thus the PI control law is fused into the neural network. The connections between neurons are improved, in order to realize the on-line adjustment of parameters, and the model does not need accurate system. The result of system simulation and experimental shows that the system using this method can track state changes quickly and adjust parameters automatically, which has strong effectiveness.

Keywords: Particle swarm optimization, neural network, photovoltaic, inverter.