MVC-based performance assessment with external hybrid disturbances

Abstract

Control performance assessment (CPA) is a useful tool to estimate the quality of industrial feedback control loops. Most MVC-based CPA techniques developed are based on the assumption of stationary disturbance, the conventional algorithms are not available for performance assessment in industrial practice, where disturbance is hybrid and is composed of periodic disturbance and stochastic noise. In this paper, a practical algorithm is developed for MVC-based CPA subjected to non-stationary disturbances. In the presence of non-stationary disturbances, the benchmark of MVC-based CPA is estimated on the basis of routine operating data without additional knowledge. The effectiveness of the proposed algorithm is demonstrated by a numerical case study.