Adaptive neuro-fuzzy modelling of anaerobic digestion of primary sedimentation sludge |
| |
Authors: | Mehmet Cakmakci |
| |
Institution: | (1) Environmental Engineering Department, Zonguldak Karaelmas University, 67100 Incivez, Zonguldak, Turkey |
| |
Abstract: | Modelling of anaerobic digestion systems is difficult because their performance is complex and varies significantly with influent
characteristics and operational conditions. In this study, Adaptive Neuro-Fuzzy Inference System (ANFIS) were used for modelling
of anaerobic digestion system of primary sludge of Kayseri municipal WasteWater Treatment Plant (WWTP). Effluent Volatile
Solid (VS) and methane yield were predicted by the ANFIS. Two stage models were performed. In the first stage, effluent VS
concentration was predicted using pH, VS concentration, flowrate of pre-thickened sludge and temperature of the influent as
input parameters. In the second stage, effluent VS concentration in addition to first stage input parameters were used as
input parameters to predict methane yield. The low Root Mean Square Error (RMSE) and high Index of agreement (IA) values were
obtained with subtractive clustering method of a first order Sugeno type inference. The model performance was evaluated with
statistical parameters. According to statistical evaluations, the models satisfactorily predict effluent VS concentration
and methane yield. |
| |
Keywords: | Anaerobic digestion Primary sludge ANFIS Model |
本文献已被 PubMed SpringerLink 等数据库收录! |
|