Ma, Keming; (2002) Process design for controllability of nonlinear systems with multiplicity. Doctoral thesis (Ph.D), UCL (University College London).

Text Process_design_for_controllabi.pdf Download (9MB)

Abstract

This thesis is concerned with the development of methods for controllability analysis leading to process design modifications of nonlinear chemical processes with input multiplicity. The first part of this thesis presents an approach to controllability analysis, based on bifurcation and continuation techniques, that can identify input multiplicity behaviour in the parameter space and give insights into the dependence of input multiplicity on the values of operating and design parameters. The algorithm developed incorporates the necessary condition for the existence of input multiplicity at a variety of steady states as an add-in subroutine into an available bifurcation analysis program, which is suitable for sizeable nonlinear processes. This allows one to study how operating conditions and design parameters influence input multiplicity behaviour, hence providing guidance to modify process designs to eliminate or avoid input multiplicity. The key features and application of the proposed approach are demonstrated through an exothermic continuous stirred tank reaction (CSTR) example and comparison made with the analytical results. The second part of this thesis focusses on an approach to making process design modifications by using optimisation and bifurcation analysis. A process modification problem is formulated within an optimisation framework which aims at minimising design parameter adjustment to eliminate potential control difficulties associated with input multiplicity behaviour for a disturbance, and results in a nonlinear programming (NLP) problem. Results are presented for its applications to a reactor-separator system with recycle and an industrial polymerization reaction.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item