ARTIFICIAL NEURAL-NETWORK-BASED LOW-DIMENSIONAL MODEL FOR SPATIO-TEMPORALLY VARYING CELLULAR FLAMES

In general obtaining a mathematical model from experimental data of a system with spatio-temporal variation is a challenging task. In this a rticle Karhunen-Loeve (KL) decomposition and artificial neural network s (ANN) are used to extract a simple and accurate dynamic model from v ideo data from experiments of two-dimensional flames of a radial extin ction mode regime. The KL decomposition is used to identify coherent s tructures or eigenfunctions of the system. Projections onto these eige nfunctions reduce the data to a small number of time series. The ANN i s then used to process these time series. As a result a low-dimensiona l, nonlinear dynamic model is obtained. (C) 1997 by Elsevier Science I nc.