OPTIMIZATION OF A CRUDE DISTILLATION UNIT USING PARAMETRIC DESIGN OF TAGUCHI AND RESPONSE SURFACE METHODS

This research work proposes a systematic procedure for determining and reducing the
number of decision variables in the case of operational optimization of a crude
distillation unit (CDU). The proposed procedure is based on the parametric design of
the Taguchi and response surface methods. The Taguchi method quantifies the
significance of variables based on their contribution factors. Variables with more than
90% contribution factors are selected and validated with the response surface method.
A sequential quadratic programming (SQP) algorithm is then employed to optimize a
profit function based onthereduced number of decision variables.