Neural networks in front-end processing and control

The authors illustrate the use of a neural network known as the multilayer perceptron as a method for solving several different tasks, all drawn from the field of tokamak research. The use of the multilayer perceptron as a nonlinear controller in a feedback loop is discussed. An outline is presented of the type of problem which can be usefully addressed by these techniques, even before the large-scale parallel processing hardware under development becomes cheaply available. Some of the difficulties encountered in applying these networks are addressed.<
>