Artificial Neural Network Modeling for Organic and Total Nitrogen Removal of Aerobic Granulation
Date
2018-06-12
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Aerobic granulation is a recent technology with high level of complexity and sensitivity to environmental and operational conditions. To understand this technology properly, mathematical modeling would be an invaluable tool. In this study, Artificial neural network (ANN), a computational tool capable of describing complex nonlinear systems, was selected to simulate the treatment performance of aerobic granulation technology. The model capability in predicting chemical oxygen demand (COD) and total nitrogen (TN) removal efficiencies of aerobic granular reactors under start-up and steady-state condition was thoroughly investigated. The capability of ANN has been examined and compared to a novel and a traditional modeling approach, namely Support Vector Machine (SVM) and Multiple Linear Regression (MLR), respectively. The models were fed with vast datasets collected from laboratory-, pilot, and fullscale studies on aerobic granulation reported in the literature. Statistical error analysis demonstrated that the ANN method yielded comparable or superior prediction accuracy, compared to other methods. The results of this study showed that ANN-based models were capable simulation approach to predict a complicated process like aerobic granulation.
Description
Keywords
Citation
Gong, H. (2018). Artificial Neural Network Modeling for Organic and Total Nitrogen Removal of Aerobic Granulation (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/31992