A comparison of nonlinear control performance assessment techniques for nonlinear processes

Abstract

Assessing the quality of industrial control loops is an important auditing task for the control engineer. However there are complications when considering the ubiquitous nonlinearities present in many industrial control loops. If one simply ignores these nonlinearities, there is the danger of over-estimating the performance of the control loop in rejecting disturbances, and thereby possibly overlooking loops that need attention. To address this problem several techniques have been recently developed to extend the control performance assessment (CPA) of single input/single output linear systems to nonlinear systems. This paper surveys these nonlinear CPA techniques and compares them using three case studies. These results can be used to guide control engineers to select the most suitable CPA techniques for their particular applications.

Minimum variance lower bound estimation with Gaussian Process models

“…Generally, there are three techniques for estimating the MVLB for nonlinear systems: a linear technique, a parametric method and an ANOVA-based method. The first method, primarily included as a benchmark, is the standard linear approach where one simply ignores nonlinearity; the second approach is a parametric method originally proposed to deal with nonlinear systems with additive disturbances (Harris and Yu, 2007; Prabhu, 2008; Zhang et al, 2011; Zhou and Wan, 2008); and the third method is based on the technique of an ANOVA-like variance decomposition (Harris and Yu, 2007; Prabhu, 2008; Yu et al, 2010; Yu et al, 2012; Zhang et al, 2011; Zhou and Wan, 2008).…”

“…Method 3 can be used to assess the control loop performance in cases where the control invariant does not exist. Nevertheless, Method 3 demands intensive computation and utilizes the Monte-Carlo method (Yu et al, 2012).…”