Fuzzy modeling system based on GA fuzzy rule extraction and hybrid of differential evolution and Tabu search approaches: Application in synthesis gas conversion to valuable hydrocarbons process. (1st March 2023)

• Record Type: Journal Article
• Authors: Tayyebi, Shokoufe; Hajjar, Zeinab; Soltanali, Saeed
• Is Part Of: Fuel. Volume 335(2023)
• Publication Date: 2023-03-01
• Subjects: Fuzzy modeling system -- Fuzzy rule extraction -- Synthesis gas conversion to hydrocarbon process -- Genetic algorithm -- Tabu search optimization -- Differential elevation; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2022.126980
• Languages: English
• ISSNs: 0016-2361

Abstract

Graphical abstract: Highlights: GA Fuzzy approach is proposed to model synthesis gas conversion to hydrocarbons process. Combination of DE and Tabu search techniques is proposed to optimize fuzzy set parameters. Fuzzy rule extraction based on K-means clustering is applied. A logarithmic distance is proposed to provide an initial guess for clusters and their centers. The prediction of CO conversion and olefin and paraffin selectivities is performed by the proposed method. Abstract: In the present study, a fuzzy approach is employed in order to develop a model for the conversion of synthesis gas to valuable hydrocarbon process. The fuzzy rules are extracted by clustering technique in this study. The numbers of fuzzy sets for input and output variables are optimized by genetic algorithm (GA) so that the number of data which follows the inconsistent rules, is minimized. Firstly, the center values of clusters are estimated by a proposed logarithmic distance method and the clusters and their centers are then calculated by k-means method. The prediction of CO conversion and olefin and paraffin selectivities is performed based on five input variables: H2/CO content, Time on Stream (TOS), temperature, Weight Hourly Space Velocity (WHSV) and phosphorous content by the proposed fuzzy model so that the parameters of which are optimized by combination of Differential Elevation and Tabu Search optimization approaches (DETS). The comparison of the performance of the designed approaches with two other models (i.e. fuzzy and GA fuzzy models) is executed and the results clearly demonstrate that the developed DETS- GA clustering fuzzy model can be defined as a powerful tool in modeling of the conversion of synthesis gas to hydrocarbons process with regression analysis (R 2 ) of over 96%.