A Decision Support System for consumption optimization in a naphtha reforming plant

In a naphtha distillation process, the natural objective is to perform an entire process maximization of the production rate while meeting required product qualities by searching for an optimal operating condition by manipulating the operating variables. The objective of this paper includes performing an energy process optimization. Not only is an adequate production rate met with the required product qualities but the operating cost is also minimized through a data mining approach. The study of the influence of all process attributes in the defined Energy Efficient Indicator (EEI) allows the construction of a multivariate linear model to aid human experts in the recovery of energy losses. A canonical discriminant function carried out the data prediction step. The quality of the Decision Support System framework is illustrated by a case study considering a real database. Also, a commercial software supported by this mining framework is presented.

► We present a data mining framework to improve a naphtha reforming plant efficiency.
► The framework presented uses only the historical database information.
► We suggest a practical way for operating cost minimization.
► We present a real case study and a framework-based commercial software.