Evaluating the performance of a sequencing batch reactor for sanitary wastewater treatment using artificial neural network

Abstract

In this work, the performance of a sequencing batch reactor (SBR) was studied for treating sanitary wastewater of Yazd power plant, Iran. For this purpose, at the first, a pilot system was designed, installed, and started up. Then the effects of retention time, pH, temperature, influent chemical oxygen demand (COD) concentration, and air flow rate were investigated on the effluent concentration of COD. In SBR reactor used in the Yazd power plant, the microalga was not used for the wastewater treatment. In this case, the COD effluent output was, at the best conditions, approximately 92 mg/L. In the studied SBR system, we used the Chlorella vulgaris microalgae and microorganisms, simultaneously. In this case, the COD level reached 34 mg/L. An artificial neural network (ANN) was developed by applying Levenberg-Marquardt training algorithm to predict the effluent concentration of COD. The optimum conditions were obtained at pH = 8, temperature of 30 C, influent COD concentration of 600 mg/L, and air flow rate of 50 L/min. ANN predicted results were in good agreement with the experimental data with a validation coefficient of determination (R 2) and validation mean square error of 0.962 and 0.0015, respectively.

Modelling selenite and chemical oxygen demand removal in an inverse fluidized bed bioreactor using genetic algorithm optimized artificial neural network

“…31,32 The ANN-based modelling approach has been used to predict chemical oxygen demand (COD) removal as a function of HRT, temperature, pH, influent COD load, and airflow in SBRs treating sanitary wastewater. 33 In autotrophic denitrification using FBR, multilayer perception feed-forward ANN has been used to model thiosulfate (S 2 O 3 2− ), and nitrate removal as a function of influent concentrations of contaminants, pH, and DO. 34 The organic loading rate, and retention time (along with total dissolved solids) have been used to model effluent COD for hypersaline oily wastewater treatment in a membrane SBR, using a feed-forward neural network.…”

“…Data‐driven black‐box model, e.g., ANN requires no formal mathematical relationship and can be of great use in the absence of detailed mechanistic insight into complex biological processes 31,32 . The ANN‐based modelling approach has been used to predict chemical oxygen demand (COD) removal as a function of HRT, temperature, pH, influent COD load, and airflow in SBRs treating sanitary wastewater 33 . In autotrophic denitrification using FBR, multilayer perception feed‐forward ANN has been used to model thiosulfate (S 2 O 3 2− ), and nitrate removal as a function of influent concentrations of contaminants, pH, and DO 34 .…”

Modeling the degradation of polyamide-12 by biocompatible fuels using artificial neural networks

“…Second, the network makes no assumptions about data distribution, and third, neural networks are very flexible about incomplete and missing data. [12][13][14][15][16][17][18] Also, neural networks have been successful in solving and identifying complex patterns and classification problems and have been used in different fields and have obtained acceptable results. Therefore, due to the satisfactory network history in predicting a large range of parameters in various sciences, it is expected that the degradation temperature of polyamide-12 can be well predicted by a neural network.…”