Control performance assessment based on sensor fusion techniques

• A control performance assessment technique for multi-loop control systems is presented.
• Decisions from different monitoring algorithms are fused based on Bayesian and Dempster–Shafer theory.
• Both of the Bayesian and Dempster–Shafer theories have their pros and cons, therefore, decisions from these both methods are fused to make final decision.

Control performance assessment techniques are widely studied and many performance assessment indices have been proposed. In this paper, a control performance assessment technique for multi-loop control systems is presented based on the decision fusion strategy. Since decisions based on individual indices can lead to erroneous results, decision fusion of different indices can improve the assessment accuracy, especially in multi-loop control systems in the presence of loop interactions. Performance assessment indices are individually evaluated and decisions based on these indices are fused. The results of simulation and practical implementation on series cascade control structures illustrate the effectiveness of the proposed algorithm.