Hybrid genetic design of GMDH-type neural networks using singular value decomposition for modelling and prediction of the explosive cutting process
Nader Nariman-zadeh, A. Darvizeh, G. R. Ahmad-zadeh · Proceedings of the Institution of Mechanical Engineers Part B Journal of Engineering Manufacture · 2003
A genetic algorithm and singular value decomposition (SVD) are deployed simultaneously for optimal design of both connectivity configuration and the values of coefficients respectively involved in group method of data handling (GMDH)-type neural networks, which are used for modelling the explosive cutting process of plates by shaped charges. The aim of such modelling is to show how the depth of penetration varies with the variation in important parameters, namely the apex angle, standoff, liner thickness and mass of charge. It is also demonstrated that SVD can be effectively used to find the vector of coefficiencs of quadratic subexpressions embodied in such GMDH-type networks. Such application of SVD will improve the performance of evolved GMDH-type networks to model the very complex process of explosive cutting of plates by shaped charges.