Non-linear Control of Parallel AC Voltage Source Converter High-Voltage DC System Using a Hybrid Input-output Linearization Evolutionary Fuzzy Controller

Abstract

This article develops a continuous-time state-space model for a parallel AC voltage source converter high-voltage DC system in the d-q reference frame. A non-linear input-output linearization fuzzy proportional-integral type controller has been designed using just two rules with a view to implement in real time for the high-voltage DC light system based on two insulated-gate bipolar transistor based converters and a pulse width modulation firing scheme. The input-output linearization transforms the non-linear voltage source converter high-voltage DC converter system into one that is linear for obtaining the control inputs. However, the uncertainty due to speed and load fluctuations, along with converter parameter variations, limits the efficacy of the input-output linearization controller for enhancing the stability of the parallel AC voltage source converter high-voltage DC system. Thus, an analytical fuzzy controller for the converter-inverter system via feedback linearization with input-output decoupling is proposed for improved damping performance. The analytical fuzzy controller is then optimized using a particle swarm optimization technique to provide superior damping performance over a wide range of operating conditions.