Use of Neural Nets for Determining the Spinnability of Fibres

It is very important for a spinner to be able to predict the degree of spinnability of a given fibre quality. Certain process conditions must be taken into account here. This paper describes how the spinnability or a given fibre quality on a rotor and ring spinning machine can be predicted with a reliability of 95% by means of a neural network. The structure and the characteristics or the neural net used will be considered in greater depth, and a simple method of implementation of such a neural net will be described.