Conventional mechanics of cutting approach for prediction of thrust and torque in drilling makes use of the oblique cutting theory and orthogonal cutting data bank.The quantitative reliability,in these models,depends on the 'input parameters' along with the 'edge force' components from the orthogonal cutting data bank for that given work material.By contrast,neural networks for drilling performance predication have been shown to be successful for quantitative predications with minimum number of inputs.In this paper neural network architecture is proposed which uses process variables such as tool geometry and operating conditions to estimate thrust and torque in drilling.Extensive drilling tests are carried out to train the feed forward back propagation network with multiple layers.The developed network i tested over a range of process variables to estimate thrust and torque.It is shown in this work that using the neural network architecture the drilling forces are 'simultaneously' predicted within 5
展开▼
