Soft-sensing estimation of plant effluent concentrations in a biological wastewater treatment plant using an optimal neural network

作者：

Highlights：

• We develop a neural-network-based soft sensor to predict effluent concentrations.

• We estimate primary variables from secondary variables in activated sludge process.

• Principal component analysis is applied to optimal selection of sensor input vector.

• The soft sensor reduces investment and maintenance costs associated with monitoring.

摘要

•We develop a neural-network-based soft sensor to predict effluent concentrations.•We estimate primary variables from secondary variables in activated sludge process.•Principal component analysis is applied to optimal selection of sensor input vector.•The soft sensor reduces investment and maintenance costs associated with monitoring.

论文关键词：Neural networks,Optimal soft sensing,PCA selection,Activated sludge process,Effluent concentrations,GPS-X simulator

论文评审过程：Received 26 February 2015, Revised 13 June 2016, Accepted 14 June 2016, Available online 20 June 2016, Version of Record 30 June 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.06.028