NEURAL EVALUATION OF FRICTION AND FLOW-STRESS ADAPTIVE TO RING GEOMETRY

Interfacial friction and material flow stress can be evaluated through the use of calibration curves in ring compression testing. In this st udy the neural network approach has been extended to their evaluation adaptive to ring geometries of wider range. The ring geometries covere d were in the range of 6 : 3 : 0.5 to 6 : 3 : 2 (OD : ID : T-0), which are the most commonly used values. Data for training the networks wer e acquired in the same way as in the development of the calibration cu rves. A serial scheme for the evaluation was found to be effective whe n multilayered BP (backpropagation) networks were employed. Network co nstruction, network training including the selection of learning param eters, and implementation of the trained network are also detailed in this paper. Predictions for different ring geometric. and friction fac tors were conducted and satisfactory results Mere obtained with predic tion error of about 5%, at maximum, for both friction and flow stress.