Copyright (c) 2008 Berkeley Electronic Press All rights reserved. http://www.bepress.com/cppm Recent documents in Chemical Product and Process Modeling en-us Sat, 10 May 2008 02:30:12 PDT 3600 http://www.bepress.com/cppm/vol3/iss1/20 http://www.bepress.com/cppm/vol3/iss1/20 Thu, 08 May 2008 10:07:20 PDT This paper discusses the design and application of a filter based on an Artificial Neural Network (ANN) in a chemical engineering process. The design of a filter consists of adapting the algorithms that make up the filter to the process to be filtered. Taking into account that the ANN is able to model almost every type of chemical process, the design and application of a filter based on ANN was studied. In this work, every ANN used was based on Multilayer Perceptron (MLP). Bearing in mind that ANN should reproduce the process as accurately as possible, an optimisation of the ANN (training function and parameters) was carried out. A mathematical model of a reflux in the upper part of a distillation column was used to test the ANN filter. The ANN is able to filter noisy signals with a mean prediction error less than 2.510-3 %. Jose S. Torrecilla Process modeling Simulation and optimization http://www.bepress.com/cppm/vol3/iss1/19 http://www.bepress.com/cppm/vol3/iss1/19 Tue, 06 May 2008 19:46:40 PDT Flows in highly heterogeneous porous media are found in many practical fields, such as hydrology, petroleum engineering, and chemical engineering. The case of two-region heterogeneous media (fractured media, catalytic beds, etc.) plays a fundamental role. The different questions associated with this specific case are illustrated in this paper for two different kinds of transport: (i) flow of a slightly compressible fluid, (ii) dispersion of a tracer. Many different theoretical models are implemented using COMSOL Multiphysics. These models correspond to direct simulation and macro-scale or large-scale models such as fully averaged models or mixed models. Debenest Gérald porous media