Compensating for disturbances in control systems

Demonstrating the practical benefits of a simple method

Abstract

Control design engineers prefer precise models of a plant because these increase the chances to successfully transfer the design from simulation to reality. Furthermore, if the influence of disturbances, the noise content of the signals, and the model inaccuracy are small enough, then an open loop control might be sufficient. Otherwise, a feedback control loop is required to reduce the impact of the disturbances on the controlled variable. Using the method presented here, the time needed to reduce the impact during the process can be shortened significantly by directly estimating and compensating for the effects
caused by disturbances and noise before these become noticeable in the controlled variable.