Estimation of cutting torque in drilling system based on flexible neural network

The use of conventional neural network in complex control applications has been limited due to its structure of low degree of freedom. In this paper, a flexible neural network (FNN), which gives flexibility of activation function, is introduced to improve the ability of the neural network, and a learning algorithm is derived. In order to show the effectiveness, the proposed method is applied to the estimation of cutting torque in a drilling process, which is a complicated control problem.