Prediction of the peak shear strength of the rock joints with artificial neural networks

With the development of computer technology, artificial neural networks are becoming increasingly useful in field engineering geology and geotechnics. networks, geomechanical properties rocks or their behaviour could be predicted under different stress conditions. Slope failures underground excavations mostly occurred through joints, which essential for stability geotechnical structures. This is why peak shear strength a rock joint most important parameter mass stability. Testing characteristics joints often time consuming suitable specimens testing difficult to obtain during research phase. The roughness surface, tensile vertical load have great influence on joint. In presented paper, surface was measured with photogrammetric scanner, determined by Robertson direct test. Based six input network, using backpropagation learning algorithm, successfully learned predict trained network similar lithological geological conditions average estimation error 6 %. results calculation were compared Grasselli experimental model, showed higher comparison model.