Abstract:Owing to the extremely complex compositions and origins of waste-activated sludge (WAS), the multiple physiochemical properties of WAS have impacts on its dewaterability, and there is a complex interaction relationship among the multiple physiochemical properties, which makes it difficult to identify the controlling factors on WAS dewaterability. Accordingly, there is still no unified certainty in the appropriate ranges of physiochemical properties for the optimal dewaterability of sludge from different sources, resulting in a lack of clear theoretical basis for technical selection and optimization of sludge dewatering processes. The large consumption of conditioning chemicals and low process efficiency stand for the major deficiency of existing sludge conditioning technologies. This study proposed to use a non-linear, adaptive and self-organizing artificial neural network (ANN) model to integrate the multiple physiochemical properties of WAS affecting its dewaterability, and WAS dewatering performance under certain conditioning schemes could be predicated by ANN model with the multiple physiochemical properties and conditioning operation parameters as the input arguments. Thus, the laborious filtration experiments for screening conditioning chemicals could be replaced by the input adjustment of ANN model. Rooted mean squared error (RMSE) of 6.51 and coefficient of determination (R<sup>2</sup>) of 0.73 confirmed the satisfied stability and accuracy of established ANN model. Furthermore, the predictor-exclusive method revealed that the exclusion of polar interface free energy decreased most, which reflected the importance of surface hydrophilicity reduction in sludge dewaterability improvement. All the contributions presented here were believed to provide an intelligent insight to improve the experience operation status of WAS dewatering process.

Development of Integration Prediction Model for Alumina Raw Slurry Quality

Raw Mix Slurry Quality Prediction Model Used for Alumina Sintering Production Process

Modeling and Optimal-Setting Control of Blending Process in a Metallurgical Industry

Blending Expert System for Raw Mix Slurry in Production of Alumina with Sintering Process

Prediction and Quantifying Parameter Importance in Simultaneous Anaerobic Sulfide and Nitrate Removal Process Using Artificial Neural Network

A Self-Training-based Approach for Aluminum Alloy Casting Quality Prediction

A Two-Stage Intelligent Optimization System for the Raw Slurry Preparing Process of Alumina Sintering Production

Real-time Optimization under Uncertainty for the Blending Process of Alumina Production and Its Application

Real-time Prediction and Simulation of Returned Materials in Blending Process of Alumina Production

The Intelligent-Integrated Model for L/S Prediction

Development and application of intelligence control optimization method for alumina batching process

Intelligent predictive modeling of blast furnace system

An Integrated Model For Caustic Ratio Prediction In The Alumina Digestion Process

Alumina Evaporation Concentration Prediction Based on Adaptive Weighted LS-SVR

Intelligent Optimization of Raw Material Blending for Alumina Production with Information Uncertainty

A Model Study on Raw Material Chemical Composition to Predict Sinter Quality Based on GA-RNN

Integrated Prediction Model of Bauxite Concentrate Grade Based on Distributed Machine Vision

Intelligent upgrade of waste-activated sludge dewatering process based on artificial neural network model: Core influential factor identification and non-experimental prediction of sludge dewatering performance

Intelligent Prediction Model of Matte Grade in Copper Flash Smelting Process

Online hybrid modeling method with application for predicting Bauxite production indicators

Moving Time Domain Blending Operational Strategy for Alumina Production by Mixed Integer Optimization