Optimisation of an Amplidyne Speed Control System

Abstract

The optimal-linear-regulator theory is used to optimise an amplidyne speed control system subjected to a constant load disturbance. As such a regulatory system tends to a steady state in response to a step disturbance. To be able to use a quadratic performance index as an optimisation criterion, three different techniques of variable transformation which moved the steady state to the origin of co-ordinates were employed. The first modification technique yielded an optimal control proportional to the state and to the load disturbance, ie, Feedforward Control. The second method led to an optimal control proportional to the state and the time-integral of the speed error, ie, Integral Feedback Control, while the third resulted in an optimal control proportional to the state and to both the load disturbance and to the time-integral of the speed error. This third control may be described as Integral Feedback/Feedforward Control. The transient responses of the system in the three cases to a step load disturbance were calculated, and the Integral Feedback/Feedforward Control gave the best results.