Multi-input variable structure controller for electronic power converters implemented by neural network techniques

A neural controller implementing a variable structure control (VSC) algorithm as a control law is proposed as an alternative to control power converters. VSC control laws yield fast response and a robust behavior against large parameters variations. An analog neural network (ANN) learns to approximate this control law by means of a learning algorithm. This implementation possesses all the good properties of sliding mode while avoiding the unnecessary discontinuities of the control input and thus eliminating chattering. A simple analog electronic implementation of this controller is proposed and applied to a PWM DC/DC power converter circuit with power factor corrector. Simulation results are summarized confirming the validity of the neural network approach