Predictive Control of a Wastewater Treatment Process

The paper deals with the design of a predictive controller for a wastewater treatment process. In the considered process, the wastewater is treated in order to obtain an effluent having the substrate concentration within the standard limits established by law (below 20 mg/l). This goal is achieved by controlling the concentration of dissolved oxygen to a certain value. The predictive controller uses a neural network as internal model of the process and alters the dilution rate in order to fulfill the control objective. This control strategy offers various possibilities for the control law adjustment by means of the following parameters: the prediction horizon, the control horizon, the weights of the error and the command. The predictive control structure has been tested in three functioning regimes, considered essential due to the frequency of their occurrence in current practice.

[1]  Arthur M. Buswell,et al.  Microbiology and Theory of Activated Sludge , 1923 .

[2]  Luis F. Diaz.  Control and Instrumentation for Wastewater Treatment Plants , 2000 .

[3]  S. A. Manesis,et al.  Intelligent control of wastewater treatment plants , 1998, Artif. Intell. Eng..

[4]  Mogens Henze,et al.  Activated sludge models ASM1, ASM2, ASM2d and ASM3 , 2015 .

[5]  E. Mosca.  MODEL-BASED PREDICTIVE CONTROL , 2006 .

[6]  G Langergraber,et al.  Monitoring of a paper mill wastewater treatment plant using UV/VIS spectroscopy. , 2004, Water science and technology : a journal of the International Association on Water Pollution Research.

[7]  Michael Johnson,et al.  Control and instrumentation for wastewater treatment plants , 1999 .

[8]  Marko Bacic,et al.  Model predictive control , 2003 .

[9]  Fatiha Nejjari,et al.  Non-linear multivariable adaptive control of an activated sludge wastewater treatment process , 1999 .

[10]  E Choi,et al.  High strength nitrogen removal from nightsoil and piggery wastes. , 2004, Water science and technology : a journal of the International Association on Water Pollution Research.

[11]  England Aj,et al.  A unified model of the activated sludge process. , 1974 .

[12]  Zvi Boger,et al.  Application of neural networks to water and wastewater treatment plant operation , 1992 .

[13]  Mietek A. Brdys,et al.  Robust Hierarchical Optimising Control of Municipal Wastewater Treatment Plants , 2001 .

[14]  H Kroiss,et al.  What is the potential for utilizing the resources in sludge? , 2004, Water science and technology : a journal of the International Association on Water Pollution Research.

[15]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[16]  Ulf Jeppsson,et al.  Modelling Aspects of Wastewater Treatment Processes , 1996 .

[17]  A. Englande,et al.  A unified model of the activated sludge process. , 1974, Journal - Water Pollution Control Federation.

[18]  Pernille Ingildsen,et al.  Realising full-scale control in wastewater treatment systems using in situ nutrient sensors , 2002 .