Controller performance assessment by frequency domain techniques

A system identification based method for assessing the performance of closed-loop systems is proposed, utilizing measures which coincide naturally with classical and modem frequency domain design specifications. Standard robust control system design methodologies seek to maximize closed-loop performance, subject to strict robustness requirements and include specifications for bandwidth and peak magnitude of the sensitivity and complementary sensitivity functions. Estimates of these transfer functions can be obtained by exciting the reference input with a zero mean, pseudo random binary sequence, observing the process output and error response, and developing a closed-loop model. Performance assessment is based on the comparison between the observed frequency response characteristics and the design specifications. Selection of appropriate model structures, experiment design, and model validation which will ensure reasonable estimates of the closed-loop transfer functions are considered in this paper. A case study involving the performance assessment of a packed bed tubular reactor control system is presented.