Environmental sensors are utilized to collect real-time data that can be viewed and interpreted using a visual format supported by a server. Machine learning (ML) methods, on the other hand, are excellent in statistically evaluating complicated nonlinear systems to assist in modeling and prediction. Moreover, it is important to implement precise online monitoring of complex nonlinear wastewater treatment plants to increase stability. Thus, in this study, a novel modeling approach based on ML methods is suggested that can predict the effluent concentration of total nitrogen (TNeff) a few hours ahead. The method consists of different ML algorithms in the training stage, and the best selected models are concatenated in the prediction stage. Recursive feature elimination is utilized to reduce overfitting and the curse of dimensionality by finding and eliminating irrelevant features and identifying the optimal subset of features. Performance indicators suggested that the multi-attention-based recurrent neural network and partial least squares had the highest accurate prediction performance, representing a 41% improvement over other ML methods. Then, the proposed method was assessed to predict the effluent concentration with multistep prediction horizons. It predicted 1-h ahead TNeff with a 98.1% accuracy rate, whereas 3-h ahead effluent TN was predicted with a 96.3% accuracy rate.
A parameter calibration of activated sludge models (ASMs) was performed to predict effluent chemical oxygen demand (COD), total nitrogen (TN), total phosphate (TP) and total suspended solid (TSS) concentrations of the enhanced biological phosphate removal process. Such calibration is an essential process for simulating the behavior of real-world wastewater treatment processes properly. Six different simulations were attempted to develop a reliable calibration method using two different parameter estimation methods for three objective functions. For the parameter estimation method, dynamic parameter estimation (DPE) and static parameter estimation (SPE) were investigated. The objective functions were based on the effluent quality (EQ) index of benchmark simulation and the effluent quality standards (EQS) in Korea. When using the same parameter estimation method, the predicted errors with the EQS-based objective functions could be decreased by approximately 20% over EQ index-based functions for TSS. When using the same objective function, the error with DPE was around 40% less than the error with SPE for TSS and TP. Therefore, applying DPE to the objective function based on the EQS was a proper calibration method of the ASM to predict a reliable effluent for the real process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.