Artificial neural network for modelling thermal decompositions

This paper introduces a new approach based on artificial neural network (ANN) to study the kinetics of thermal decomposition reactions of different polymeric materials, using dynamic thermogravimetry analysis (TGA) at several heating rates. A multilayer neural network model was trained and tested using published experimental data, allowing the proposed model a very good correlation of the weight-loss data. As an example, the same kinetic model has been applied to simultaneously fit runs performed at different heating rates, with different polymeric materials such as polyethylene, cellulose and lignin. (C) 2003 Elsevier B.V. All rights reserved.