Tool Wear Estimation in Face Milling Using Spindle Motor Parameters with Neural Networks

The monitoring of tool wear is one of the important methods to improve the dimensional accuracy and economic aspects of machining. To achieve this, instead of measuring the tool wear, other parameteres, which are related to tool wear are used. In this paper, an intelligent system, based on neural networks, is presented. By this method, the tool wear is estimated on-line with measuring spindle motor parameters, such as current and the speed of motor. Thus, the current and the speed of motor in different condition of machining (feed, depth of cut and rpm of tool) and wear were measured with practical experiments and the effects of tool wear on current and speed of motor is analized. Based on the results, a back propagation (BP) neural network is developed and trained. Using this network, the tool wear could be estimated while machining in different conditions, measuring current and speed of motor. This system could be used to control and monitor the machining process.