StatCom´s Voltage Regulation by a Neurocontroller

This paper is aimed to the control of the voltage magnitude of a static synchronous compensator (STATCOM´s) prototype at lab-scale. This is based on the pulse width modulation (PWM) technique. A B-Spline neural network (B-SNN) control strategy is proposed for such purposes, and has been selected due to its robustness, simple structure, adaptability, and fast response. This neuro-controller possesses generalization characteristics that compare favorably with other neural networks such as the backpropagation and radial basis function networks, respect to the learning rule convergence. Laboratory tests demonstrate the applicability of the proposition.