Taguchi’s Parametric Design Approach for Determining the Significant Decision Variables in a Condensate Fractionation Unit (CFU)

Taguchi’s parametric design approach is employed to systematically determine the significant decision variables and their respective optimal levels contributing towards the most profitable operation of a Condensate Fractionation Unit (CFU). The present work aims to reduce the frequent optimization issues such as high market dynamics in term of fluctuation of price of condensate feedstock and products as well as the difficulties in determining the decisive variables for effective implementation of optimization strategy. For this purpose, a steady-state CFU model developed under HYSYS environment is used as a virtual plant to carry out the fractionation processes. Experiments are designed and executed by utilizing the combination of process parameters (9 controllable and 2 noise factors) based on three-level of and orthogonal arrays. The results are interpreted using analytical tools such as analysis of mean (ANOM), analysis of variance (ANOVA), signal-to-noise ratio (SNR) analysis and response plot. Five controllable factors handling the pressure, temperature and product flow rates top the ranked list with a total contribution of 98.2%. The maximum profit is obtained from an optimal configuration of both controllable and noise factors. Remarkable agreements with an average deviation of 0.45% are found in all cases when the implemented results are validated against those suggested by ANOM and the improved profit further verifies the optimality of configuration. The outcome from this work imply that Taguchi method can be employed in other processes due to its robustness in handling noise factors with the minimum number of experiments.