Performance Evaluation of Generalized Minimum Variance Multi-disturbance Control System

The minimum variance benchmark is too radical, because it only pursues the minimum variance without considering the robustness of control system, resulting in the life of equipment reducing. So other advanced benchmarks are derived. Quadratic Gaussian benchmark and generalized minimum variance benchmark are directing extensions of the minimum variance benchmark. The penalty for the control effect is considered to avoid the phenomenon of extreme minimum variance. The robustness of the system is greatly improved and the evaluation effect is better. Industrial processes often encounter multi-disturbance dynamics, such as the simultaneous existence of tracking disturbance and random disturbance, or the simultaneous existence of two different kinds of disturbance between actuator and controlled object. Therefore, in this paper, the performance evaluation of generalized minimum variance multi-disturbance control system is proposed.