Controllability analysis and modelling requirements: an industrial example

Abstract

This paper presents a controllability analysis procedure and modelling approach developed to address problems raised by industrial applications of controllability research. Most work on controllability analysis assumes a nonlinear dynamic model as its starting point using linearisation of the model to give a model for frequency domain analysis. Such models may be expensive and difficult to develop. This paper looks at the use of low order empirical dynamic models for controllability analysis. A heuristic controllability analysis procedure is presented which lends itself to the use of simple models and addresses the use of cascade controllers, which are a feature of many industrial problems, in a systematic way. An industrial example is presented and the procedure is applied successfully using simple gain-delay-lag models. © 1995.