Sequencing batch reactor (SBR) is a type of activated sludge treatment, in which the aeration and clarifying step take place in a single reactor. There are several literatures and patents that have investigated the ability of SBR in wastewater treatment. In this study, the short-time effects of operative parameters, including influent COD, COD/N ratio, COD/P ratio, aeration time, settling time, temperature, and the memory of reactor on performance of a SBR in treating starchy wastewater were investigated. Fractional factorial design (FFD) approach was applied for the design of experiment. Fuzzy inference systems (FIS) with Mamdani's and Sugino's method were developed to simulate the effects of operative parameters. Accordingly, the ability of several structures of FIS was investigated. There was a significant difference between Mamdani's and Sugino's model with wtaver defuzzification. But the model accuracy of Sugino's with wtsum is very poor. Generally, the ability of Mamdani' model in response prediction is higher than Sugino's. The correlation coefficient index (R2) of the best FIS structure was 0.9774.

Keywords