Design of longwall face support by use of neural network models

The relationship between support performance parameters and parameters representative of mining and roof conditions is characterised by uncertainty, non-linearity and dependence, so the theory of artificial neural networks was selected as a basis for shield design. A computer program based on a back-propagation training algorithm was written to derive the weighting factors between inputs and outputs from field data; once these had been established two mathematical models were developed to describe the relationship. Back-calculation of field data indicates that the results generated from the neural network models are much more accurate than those derived by traditional methods and can be used with a higher level of confidence. The support density, possible yield frequency of face supports and interval of periodic roof weighting under specific mining and roof conditions can be determined by application of the models