Artificial neural network modelling for organic and total nitrogen removal of aerobic granulation under steady-state condition

Modelling traffic noise in a wide gradient interval using artificial neural networks

As classical traffic noise prediction models lack a deeper consideration of the impact of the gradient, the characteristics of longitudinal gradients from multiple roads were collected as data in the mountain city of Chongqing county, which was chosen as the entry point, to study the noise characteristics for a wide range of road gradients and to build a traffic noise prediction model based on artificial neural networks (ANNs). The field data consisted of traffic volumes, heavy-vehicle ratios, average vehicle speeds, road gradients, and corresponding equivalent sound pressure levels. An optimal ANN model was determined and compared with two classical models. The results demonstrated that a one-hidden-layer ANN model was suitable for traffic noise prediction in mountain cities and presented better predictive performance than the conventional models. The best-performing ANN model yielded a determination coefficient of 0.9447 and a mean-squared error of 0.2708 dBA. Moreover, this study confirmed that road gradients were significant for constructing traffic noise prediction models.

Support Vector Regression Modelling of an Aerobic Granular Sludge in Sequential Batch Reactor

Support vector regression (SVR) models have been designed to predict the concentration of chemical oxygen demand in sequential batch reactors under high temperatures. The complex internal interaction between the sludge characteristics and their influent were used to develop the models. The prediction becomes harder when dealing with a limited dataset due to the limitation of the experimental works. A radial basis function algorithm with selected kernel parameters of cost and gamma was used to developed SVR models. The kernel parameters were selected by using a grid search method and were further optimized by using particle swarm optimization and genetic algorithm. The SVR models were then compared with an artificial neural network. The prediction results R² were within >90% for all predicted concentration of COD. The results showed the potential of SVR for simulating the complex aerobic granulation process and providing an excellent tool to help predict the behaviour in aerobic granular reactors of wastewater treatment.

Development of an ensemble of machine learning algorithms to model aerobic granular sludge reactors

Machine learning models provide an adaptive tool to predict the performance of treatment reactors under varying operational and influent conditions. Aerobic granular sludge (AGS) is still an emerging technology and does not have a long history of full-scale application. There is, therefore, a scarcity of long-term data in this field, which impacted the development of data-driven models. In this study, a machine learning model was developed for simulating the AGS process using 475 days of data collected from three lab-based reactors. Inputs were selected based on RReliefF ranking after multicollinearity reduction. A five-stage model structure was adopted in which each parameter was predicted using separate models for the preceding parameters as inputs. An ensemble of artificial neural networks, support vector regression and adaptive neuro-fuzzy inference systems was used to improve the models' performance. The developed model was able to predict the MLSS, MLVSS, SVI₅, SVI₃₀, granule size, and effluent COD, NH₄-N, and PO₄^3- with average R², nRMSE and sMAPE of 95.7%, 0.032 and 3.7% respectively.