EXPERIMENTAL INVESTIGATION AND FUZZY LOGIC MODELING OF 8X8 CM2 MEMBRANE PERFORMANCE OF MICROBIAL FUEL CELL
- 1 Selcuk University, Department of Mechanical Engineering, Konya, Turkey
- 2 KTO Karatay University, Department of Mechanical Engineering, Konya, Turkey
In this study, microbial fuel cell’s energy conversion performance experimentally investigated from the chemical energy of the organic waste to electrical energy by means of microorganisms. Microbial fuel cell (MFC) consists of two cells which has 15x15x15 cm3 volume. One part of the cell conserves the mud (anode) the other part conserves the water (cathode). The membrane of the microbial fuel cell has 8×8 cm2 area. Two different samples were used in the experiments which are active and settlement mud. The power, volt and current values of the active and settlement mud for different temperature, resistance and bubble were determined. The temperature values consist of ΔT = 8°C, ΔT = 10°C, ΔT = 12°C, ΔT = 14°C. ΔT=Tenvironment- Tmud. For every ΔT value 2 different bubble values were examined (High=21,5 g/h, low=3,5 g/h). For every bubble effect 7 different resistance values were determined (1. Resistance= 3,75 Ω; 2. Resistance =7,5 Ω; 3. Resistance =10,5; 4. Resistance = 14,5 Ω; 5. Resistance = 16 Ω; 6. Resistance = 19 Ω; 7. Resistance = 21,5 Ω) and the performance of the 8×8 cm2 membrane of the MFC is detected. As a result; with the increase of the temperature, resistance and bubble effect the voltage production increases and correspondingly the current decreases. When all the experimental results are evaluated,the highest voltage production (687 mV) occurred at ΔT = 14°C and 21,5 Ω with the high bubble effect in the settlement mud. Also, in this study, MFCs performances in terms of voltage, current, temperature, power was modeled with Rule-Based Mamdani-Type Fuzzy (RBMTF) modeling technique. Input parameters ΔT and time; output parameter power was described by RBMTF if-the rules. 1792 experimental data sets, which obtained for power according to ΔT and time, were used in the training step. The comparison between experimental data and RBMTF is done by using coefficient of multiple determination (R2). The actual values and RBMTF results indicated that RBMTF can be successfully used in MFC.