In the petroleum and chemical engineering disciplines, numerous studies are being conducted to develop new efficient catalysts for desulfurization process. It should be noted that the development of new materials to be used in refinery units requires tremendous number of experiments to be performed in the laboratories. The conditions and parameters are to be optimized. The optimization of experimental parameters is normally costly and prolonged. That is, consuming large amount of chemicals and materials which is of cost and environmental impact and time consuming. Thus, a fast and simple optimization of parameters is highly desirable in the desulfurization context. In this research, we will investigate the application of type-2 fuzzy logic algorithms to model for modelling in oil refinery, in particular, desulfurization process. The project work will start with further updating survey on existing computational intelligence based prediction model for desulfurization process. In addition, a part of this research is to collect data samples for desulfurization of oil using nanomaterials as catalysts, and study their characteristics and significance for the prediction model. In this regard, we aim to use feature reduction techniques (e.g. Principal Component Analysis) and genetic algorithms for the sake of improving the performance of the developed model and reducing the multicollinearity. Finally, the proposed predication model will be empirically validated with respect to their prediction accuracy, required training data and training time.
|Effective start/end date
|1/09/17 → 1/08/18
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.