Neural network modeling for on-line estimation of nutrient dynamics in a sequentially-operated batch reactor.

In monitoring and controlling wastewater treatment processes, on-line information of nutrient dynamics is very important. However, these variables are determined with a significant time delay. Although the final effluent quality can be analyzed after this delay, it is often too late to make proper adjustments. In this paper, a neural network approach, a software sensor, was proposed to overcome this problem. Software sensor refers to a modeling approach inferring hard-to-measure process variables from other on-line measurable process variables. A bench-scale sequentially-operated batch reactor (SBR) used for advanced wastewater treatment (BOD plus nutrient removal) was employed to develop the neural network model. In order to improve the network performance, the structure of neural network was arranged in such a way of reflecting the change of operational conditions within a cycle. Real-time estimation of PO3-(4), NO-3, and NH+4 concentrations was successfully carried out with the on-line information of the SBR system only.

[1]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[2]  Takane Kitao,et al.  Advanced Treatment of Domestic Wastewater Using Sequencing Batch Reactor Activated Sludge Process , 1993 .

[3]  G. T. Tsao,et al.  Neural network modeling of batch cell growth pattern , 1993, Biotechnology and bioengineering.

[4]  A. E. Greenberg,et al.  Standard methods for the examination of water and wastewater : supplement to the sixteenth edition , 1988 .

[5]  A. E. Greenberg,et al.  Standard methods for the examination of water and wastewater. 14th edition. , 1976 .

[6]  James T. Luxhoj.  Neural Networks in Bioprocessing and Chemical Engineering , 1997 .

[7]  A J Morris,et al.  Enhancing bioprocess operability with generic software sensors. , 1992, Journal of biotechnology.

[8]  J Thibault,et al.  On‐line prediction of fermentation variables using neural networks , 1990, Biotechnology and bioengineering.

[9]  Y. Andersson,et al.  Waterborne Campylobacter in Sweden: the cost of an outbreak , 1997 .

[10]  S. J. Wilcox,et al.  A neural network, based on bicarbonate monitoring, to control anaerobic digestion , 1995 .

[11]  Andrea G. Capodaglio,et al.  Wastewater Treatment Plants under Transient Loading – Performance, Modelling and Control , 1993 .

[12]  Michael Nikolaou,et al.  Dynamic process modeling with recurrent neural networks , 1993 .

[13]  Cheng-Nan Chang,et al.  Monitoring and control using on-line orp on the continuous-flow activated sludge batch reactor system , 1997 .

[14]  A. J. Morris,et al.  Artificial neural networks in process engineering , 1991 .

[15]  Thomas J. McAvoy,et al.  Modeling Nutrient Dynamics in Sequencing Batch Reactor , 1997 .

[16]  O. Hao,et al.  Sequencing batch reactor system for nutrient removal : ORP and pH profiles , 1996 .

[17]  N. V. Bhat,et al.  Use of neural nets for dynamic modeling and control of chemical process systems , 1990 .

[18]  S Linko,et al.  Neural networks as 'software sensors' in enzyme production. , 1997, Journal of biotechnology.

[19]  Lloyd H. Ketchum,et al.  Design and physical features of sequencing batch reactors , 1997 .

[20]  G Stephanopoulos,et al.  Studies on on‐line bioreactor identification. I. Theory , 1984, Biotechnology and bioengineering.

[21]  Horng-Guang Leu,et al.  Effects of flow velocity and depth on the rates of reaeration and bod removal in a shallow open channel , 1997 .

[22]  Loukas Petrou,et al.  Identification and control of anaerobic digesters using adaptive, on-line trained neural networks , 1997 .

[23]  J. F. Reid,et al.  A prototype neural network supervised control system for Bacillus thuringiensis fermentations , 1994, Biotechnology and bioengineering.