J. Svensson et al., A framework for the investigation of multiparametric dependences applied to total radiated power of JET plasmas, PLASMA PHYS, 43(4), 2001, pp. 405-416
A framework is developed for investigating complex multivariate relationshi
ps in a dataset. This is based on using the universal approximation abiliti
es of a multi-layer perceptron (MLP) neural network to predict a quantity o
f interest from a large set of parameters. A measure of redundancy is deriv
ed, and used in such a way that the average influence on the predicted quan
tity from any parameter can be estimated. Input parameters can be ordered i
n terms of increasing redundancy and therefore assist in finding the most i
mportant parameters a phenomenon of interest depends upon. In spite of the
problem being multi-dimensional, the functional form of the one-to-one rela
tionship between a parameter and a quantity of interest can be visualized.
This framework is then used together with sensitivity analysis to investiga
te the dependence of the total radiated power of JET plasmas on a large num
ber of parameters, leading to the identification of a much smaller set of p
arameters to be used in an effective MLP predictor of total radiated power.