Journal of Natural Gas Chemistry 20(2011)603–610 Fuzzy model prediction of Co (III)/Al 2 O 3 catalytic behavior in Fischer-Tropsch synthesis Mohammad Ali Takassi 1 , Mahdi Koolivand Salooki 2∗ , Morteza Esfandyari 3 1. Department of Science, Petroleum College, Petroleum University of Technology, Ahvaz 6198144471, Iran; 2. Oil & Gas Processing Center of Excellence, Chemical Engineering Department, University of Tehran, 11155-4563, Tehran, Iran; 3. Department of Chemical Engineering, Faculty of Engineering, Ferdowsi University of Mashad, P.O.Box1111 Mashad, Iran [ Manuscript received December 24, 2010; revised March 12, 2011 ] Abstract The application of Co (III)/Al 2 O 3 catalyst in Fischer-Tropsch synthesis (FTS) was studied in a wide range of synthesis gas conversions and compared with Fuzzy Simulation results. Present study applies fuzzy model to predicting the product composition of CH 4 , CO 2 and CO in Fischer-Tropsch process for natural gas synthesis, in which the input vector was 4-dimension including four variables (operating pressure, operating temperature, time and CO/H 2 ratio) of 70 different experiments and the output product is a composition of CO 2 , CO and CH 4 . The Mamdani algorithm has been applied to the training of the fuzzy system and the test set was used to evaluate the performance of the system including R 2 , ARE, AARE and SD. The results demonstrated that the predicted values from the model were in good consistency with the experimental data. The work indicates how fuzzy inference system (FIS), as a promising predicting technique, would be effectively used in FTS. Key words fuzzy inference system; Fischer-Tropsch; natural gas; catalyst; Co (III); Al 2 O 3 1. Introduction Fischer-Tropsch synthesis (FTS) is an interesting and promising pathway for the conversion of synthesis gas to transportation fuels. FTS has been recognized as an important alternate technology to petroleum refining in the production of liquid fuels and chemicals from syngas derived from coal, natural gas and other carbon-containing materials [1-3]. Sev- eral metals (including Fe, Co, Ni and Ru) are considered as the most common active components for FTS catalysts, due to their high FTS activity, low cost, flexible product distribu- tion and favorable engineering characteristics [4]. Owing to high activity and long durability, cobalt-based FT catalysts are currently a choice for the conversion of syn- gas to natural gas and liquid fuels. In addition, they provide the best compromise between reduced costs and the high CO conversion. Cobalt based catalysts offer favorable C 5+ selectivity as well as low water gas shift (WGS) activity for the synthesis of liquid fuels from natural gas. Supported Co catalysts with high specific rates require the synthesis of small metal crys- tallites at high local surface densities on the support, those of supports or alloys that increase the rate per surface Co (turnover rate) [5,6,7]. The FTS and WGS reactions are as follows: CO + (1 + n/2) H 2 → CH n +H 2 O CO + H 2 O → CO 2 +H 2 where, n is the average H/C ratio of the produced hydrocar- bons. Xiong et al. [8] reported that the FTS activity and se- lectivity of cobalt based catalysts could be affected by their pore sizes. Song et al. [9] also indicated that the pore size of alumina support could significantly influence the Co 3 O 4 crys- tallite diameter, catalyst reducibility and FT activity. So in our experiment we have used Co based catalyst. Often knowledge base precise modeling methods are not suitable for the complex systems due to the lack of precise knowledge about these systems, nonlinear behavior and time varying characteristics of them. This limitation introduces a tendency to modeling complex systems based on intelligent methods, such as neural networks and fuzzy modeling. There are many researches in various fields that used these methods in nonlinear system identification. The neural networks have been applied in modeling the green house effect, simulating N 2 O emissions from a temperate grassland ecosystem, and assessing flotation experiments [10-12]. Mastorocostas et al. ∗ Corresponding author. Tel/Fax: +98-918-3532398; E-mail: Koolivand.m@nisoc. ir Copyright©2011, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. All rights reserved. doi:10.1016/S1003-9953(10)60240-X