Determining controller benefits via probabilistic optimization

doi:10.1002/acs.765

Return

Determining controller benefits via probabilistic optimization

2003-08-27

Yunfeng Zhou

J. Fraser Forbes

DOI：10.1002/acs.765

Original

Save

View PDF

Abstract

En 中文

For the most part, process control research has focussed on the synthesis and tuning of controllers, which has provided a plethora of techniques that can address virtually any application. With each new control technique, a steady stream of 'successful' application results are generated and reported. Recently, a considerable number of control researchers have turned their attention to assessing the performance of installed control systems and to the diagnosis of controller performance problems. Despite successes in the areas of controller synthesis, tuning and performance analysis, almost no research has addressed the fundamental issue of determining whether the economic performance gains that are expected accrue from a proposed process control project are sufficient to justify its execution. The work presented here proposes an optimization-based technique for calculating the expected economic performance of a given control system; a method, which is analogous to analysis of variance, for determining the expected economic benefit that will arise from a particular controller improvement effort; and a sensitivity analysis approach for determining the effect of specific assumptions on control system improvement decisions. Copyright (C) 2003 John Wiley Sons, Ltd.

Keywords:

controller benefits

stochastic optimization

sensitivity analysis

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.