Cascade Control System Performance Assessment Based On Improved Generalized Minimum Variance Benchmarking

The process’s dynamic characteristics keep changing along with time. If we can’t maintain the systems properly, the control loop will perform out of specifications gradually, which may lead to degrade performances in process operations, even the degraded performances may bring out great loss of money. This paper mostly discuss two kinds of cascade control structures-series cascade control(SCC) and parallel cascade control(PCC), which are widely used in many process industry applications. The paper focuses on the analysis of these two kinds of system performance assessment method based on minimum variance (MV) benchmarking and generalized minimum variance (GMV) benchmarking.Auto-correlation analysis provides to discover the nature of disturbances acting on the process and how they affect the system by comparing current measured value with those exhibited in the past during normal operation. This paper do some research on SCC system by using auto-correlation analysis based on MV benchmarking, and presents the great significance to detect the system’s oscillation characteristics.According to the benchmarking of GMV for conventional cascade system performance evaluation method, the former researchers pursuit of generalized minimum output variance unduly, and ignore the system loop actual output variance. It causes that the system loop output is not ideal. The paper put forward improved method for control performance assessment of series cascade control system based on GMV and MV benchmarking, and make a verification by the simulation example. The results reveal that the method can be used to achieve the evaluation effectively. As for PCC, a novel method is derived based on the MV benchmarking in this paper. The optimal controllers can be obtained directly by using the decomposition technique of Diophantine equations other than the closed loop operating data. Besides, the theoretical performance bound value can be computed simultaneously. Because the system’s performance can be assessed effectively based on GMV benchmarking, the generalized minimum variance (GMV) benchmarking for parallel cascade control system is also put forward, and the simulation results prove that the proposed method is good.