Assessment and diagnosis of feedforward/feedback control system

A sequence of statistical hypothesis procedures for assessing a feedforward/feedback control performance, diagnosing and removing its fault causes is proposed. Based on the controlled output, the current variance is contributed by the feedback only effect and the combination of feedback with feedforward effect, respectively. These effect variations are systematically conducted by a sequence of the statistic hypothesis testing and the isolation strategy to compare the current control performance and the achievable benchmark operating one. A diagnosis tree with hierarchical structure is constructed. The capability of the proposal is illustrated through a simulation case with multiple faults.