Hybrid Models in Dynamic Simulation of a Biological Water Treatment Process

In biological wastewater treatment in pulp and paper industry. A lot of process measurements are available, but measurement sets do not include sufficient information on special features of the influent nor on microbial composition of the sludge. Populations of microorganisms are highly important. Basic dynamic simulation is based on LE models. The models consist of two parts: interactions are handled with linear equations, and nonlinearities are taken into account by membership definitions. Process insight is maintained, while data-driven tuning relates the measurements to the operating areas. Genetic algorithms are well suited for linguistic equation (LE) models based on nonlinear scaling and linear interactions. Activated Sludge Models provide a basis for phenomenological modelling and can be linked to process expertise. Hybrid models with a cascade approach are needed in biological wastewater treatment to cover different operating conditions.

[1]  James J. Buckley,et al.  Universal approximators for fuzzy functions , 2000, Fuzzy Sets Syst..

[2]  Esko Juuso,et al.  Intelligent Dynamic Simulation of a Fed-Batch Enzyme Fermentation Process , 2008, Tenth International Conference on Computer Modeling and Simulation (uksim 2008).

[3]  P. Minkkinen,et al.  Partial least squares modeling of an activated sludge plant: A case study , 1997 .

[4]  Mogens Henze,et al.  Activated Sludge Model No.2d, ASM2D , 1999 .

[5]  Esko Juuso,et al.  Intelligent Modelling of a Fluidised Bed Granulator used in Production of Pharmaceuticals , 2007, Simul. Notes Eur..

[6]  Krist V. Gernaey,et al.  Activated sludge wastewater treatment plant modelling and simulation: state of the art , 2004, Environ. Model. Softw..

[7]  Esko Juuso,et al.  Dynamic simulator for dosing of water treatment chemicals , 2005 .

[8]  Yrjö Hiltunen,et al.  SOM-Based Modelling for an Activated Sludge Treatment Process , 2008, Tenth International Conference on Computer Modeling and Simulation (uksim 2008).

[9]  Esko Juuso,et al.  Fuzzy Control in Process Industry: The Linguistic Equation Approach , 1999 .

[10]  J. Buckley,et al.  Can neural nets be universal approximators for fuzzy functions? , 1997, Proceedings of 6th International Fuzzy Systems Conference.

[11]  M. Mori,et al.  Simulation of an industrial wastewater treatment plant using artificial neural networks and principal components analysis , 2002 .

[12]  Pekka Teppola,et al.  MODELING OF ACTIVATED SLUDGE PLANTS TREATMENT EFFICIENCY WITH PLSR : A PROCESS ANALYTICAL CASE STUDY , 1998 .

[13]  Mogens Henze,et al.  Wastewater and biomass characterization for the Activated Sludge Model No. 2: biological phosphorus removal , 1995 .

[14]  M. Heikkinen.  Modelling of Activated Sludge Treatment Process in a Pulp Mill Using Neural Networks , 2008 .

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

[16]  Esko Juuso,et al.  Tuning of Large-Scale Linguistic Equation (LE) Models with Genetic Algorithms , 2009, ICANNGA.

[17]  Esko Juuso,et al.  Adaptive Feedback Controller for Dosage of Water Treatment Chemicals , 2004 .

[18]  Esko Juuso,et al.  Integration of intelligent systems in development of smart adaptive systems , 2004, Int. J. Approx. Reason..

[19]  James J. Buckley,et al.  Can neural nets be universal approximators for fuzzy functions? , 1999, Fuzzy Sets Syst..