Fuzzy logic modeling of performance proton exchange membrane fuel cell with spin method coated with carbon nanotube

摘要

In this study, performance of proton exchange membrane (PEM) fuel cell was experimentally investigated and modeled with Rule-Based Mamdani-Type Fuzzy (RBMTF) modeling technique. Coating on the anode side of the membrane of PEM fuel cell was accomplished with the spin method by using carbon nanotube (CNT). This fuel cell performances at 20 degrees C, 40 degrees C, 60 degrees C were investigated experimentally and the best performance was determined and benefiting from experimental data, modeled with RBMTF method. Input parameters are; temperature (T), time (s), voltage density (V/cm(2)) and current density (A/cm(2)); output parameter power density (W/cm(2)) were described by RBMTF if-then rules. Numerical parameters of input and output variables were fuzzificated as linguistic variables: Very Very Low (L-1), Very Low (L-2), Low (L-3), Negative Medium (L-4), Medium (L-5), Positive Medium (L-6), High (L-7), Very High (L-8) and Very Very High (L-9) linguistic classes. With the linguistic variables used, 81 rules were obtained for this system. The comparison between experimental data and RBMTF is done by using statistical methods. The coefficient of multiple determination (R-2) for power density of uncoated PEM and with CNT (20 degrees C) is 98.88%, power density of 20 degrees C, 40 degrees C and 60 degrees C temperatures is 97.12%. 80 values were obtained by RBMTF technique at 20 degrees C for uncoated PEM and with CNT. During discharge for 20 degrees C uncoated PEM for experimental power density maximum 0.021 Watt/cm(2) and uncoated PEM for fuzzy model maximum 0.0205 Watt/cm(2). The actual values and RBMTF results indicated that RBMTF can be successfully used in PEM fuel cell. Performance tests of the system were not done for intermediate values which were estimated with RMBTF. 78 values at 30 degrees C and 50 degrees C which are not obtained from experimental work for power density are predicted by fuzzy logic method. (C) 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.