Energy Consumption Modeling in Activated Sludge Process Using Coupling PCA-ANFIS Approach

• Main Authors: Radia Maachou, Abdeluahab Lefkir, Abdelmalek Bermad, Soraya Abdelaziz
• Format: Article
• Language: English
• Published: Iranian Institute of Research and Development in Chemical Industries (IRDCI)-ACECR 2019-12-01
• Series: Iranian Journal of Chemistry & Chemical Engineering
• Subjects: wastewater; activated sludge; anfis modeling; pca
• Online Access: http://www.ijcce.ac.ir/article_39704_ab43036749d3f28e9576d6b6a15b0333.pdf

The main challenge in Wastewater Treatment Plants (WWTP) by activated sludge process is the reduction of the energy consumption that varies according to the pollutant load of influent. However, this energy is fundamentally used for aerators in a biological process. The modeling of energy consumption according to the decision parameters deemed necessary for good control of the active sludge process namely the removal yields of parameters pollutant such as Biological Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Suspended solids (SS) and Ammoniac (NH4+) that must meet the required standards. To achieve this objective, a coupling of two approaches, the principal components analysis (PCA) method and the Adaptatif Neural  Fuzzy Inference System (ANFIS) model was envisaged, to improve the performance of fuzzy reasoning. Indeed, PCA as a factorization tool allowing the reducing of the variable that allows the reduction of the complexity of the studied phenomenon. The neuro-fuzzy learning from the data projected on the principal axes allows the improvement of the model, both in learning and validation periods. The comparative study between ANFIS model, the regression PCA model, and the coupling PCA-ANFIS method applied to the raw data was effected. The results indicate a significant improvement in the validation criteria obtained in the coupling PCA-ANFIS model compared to the other models for the learning and validation periods. The result shows that the coupling PCA-ANFIS can be used to extract information from data and to describe the nonlinearity of complex wastewater treatment processes.