Soft measurement modeling based on Hierarchically Neural Network (HNN) for wastewater treatment - Details

Author：

Qiao, Junfei (Qiao, Junfei.) (Scholars：乔俊飞) | Ren, Donghong (Ren, Donghong.) | Han, Honggui (Han, Honggui.) (Scholars：韩红桂)

Indexed by：

EI Scopus

Abstract：

A　hierarchically　neural　network　(HNN)　is　proposed　in　this　paper.　This　HNN,　contains　two　sub-neural　networks,　is　used　to　predict　the　chemical　oxygen　demand　(COD)　and　biochemical　oxygen　demand　(BOD)　concentrations.　In　the　model　the　effluent　COD　of　wastewater　treatment　is　taken　as　the　input　of　effluent　BOD.　The　three　layered　RBF　neural　network　is　used　in　each　sub-neural　network.　The　training　algorithm　of　the　proposed　HNN　is　simplified　through　the　use　of　an　adaptive　computation　algorithm　(ACA).　Meanwhile　the　results　of　simulations　demonstrate　that　the　new　neural　network　can　predict　the　key　parameters　accurately　and　the　proposed　HNN　has　a　better　performance　than　some　other　existing　networks.　©　2012　Springer-Verlag.

Keyword：

Network layers Wastewater treatment Oxygen Effluent treatment Biochemical oxygen demand Neural networks Multilayer neural networks Effluents

Author Community：

[ 1 ] [Qiao, Junfei]College of Electronic and Control Engineering, Beijing University of Technology, Beijing, China
[ 2 ] [Ren, Donghong]College of Electronic and Control Engineering, Beijing University of Technology, Beijing, China
[ 3 ] [Han, Honggui]College of Electronic and Control Engineering, Beijing University of Technology, Beijing, China