A new insight into implementing Mamdani fuzzy inference system for dynamic process modeling: Application on flash separator fuzzy dynamic modeling. (April 2020)

• Record Type: Journal Article
• Authors: Eghbal Ahmadi, Mohammad Hosein; Royaee, Sayed Javid; Tayyebi, Shokoufe; Bozorgmehry Boozarjomehry, Ramin
• Is Part Of: Engineering applications of artificial intelligence. Volume 90(2020)
• Publication Date: 2020-04
• Subjects: Mamdani fuzzy inference -- Flash separation -- Dynamic modeling -- Heuristic-based modeling -- Rule reduction; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2020.103485
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: In this work, a novel approach to model the dynamic behavior of the flash separation process (as a main building block of non-reacting stage-wise operations) based on Mamdani Fuzzy Inference Systems is proposed. This model surmounts the need to solve various types of mathematical equations governing the system and does not require thermodynamic properties which are either not available or computationally demanding. Hence it can be easily used in dynamic simulation of multi-phase flow in distributed systems. In the proposed approach the overall model is broken into several simple sub-models based on intuitive analysis of an expert. Moreover, a new fuzzy concept, named "Linguistic Composition Variable" is introduced to represent components mole fractions of each phase as a fuzzy variable. Accordingly, large number of rules which is the main shortcoming of the Mamdani Fuzzy system is significantly reduced. The performance of the proposed dynamic model is evaluated through comparing its results against their corresponding counterparts for a flash separator of crude oil. Overall MAPE (Mean Absolute Percentage Error) values of 7.17% for the gas molar fractions, 3.06% for liquid molar fractions, 10.16% for the temperature, 0.63% for the pressure and 16.44% for the liquid level of the separator have been achieved showing that the proposed fuzzy model can effectively capture the general trends of process data of the dynamic process. Graphical abstract: Highlights: A novel approach to model the dynamic behavior of the flash separation process. Bypassing the need to solve various types of mathematical equations governing the system. A new fuzzy concept, named LCV to reduce the number of rules. Effective use of intuitive analysis of an expert to develop the model. The accuracy of more than 90% for the developed model.