J. Susai Mary,∗ M.A. Sai Balaji,∗ A. Arockia Selvakumar,∗∗ and D. Dinakaran∗∗∗
Tool wear, grey wolf optimization, neural controller, drilling process,adaptive machining
Tool wear is a vital parameter in machining that affects precision, geometry, and productivity. An adaptive control system is essential to optimize the machining parameters, which results in cost saving and improves tool life. This study presents an adaptive control system with grey wolf optimizer (GWO) and a neural controller to prolong the tool life in a drilling process. A multisensory approach with acceleration and force sensors is used for real-time prediction of tool wear. The tool wear is modelled by multiple linear regression (MLR) with a regression of 0.8 and a root mean square error (RMSE) of 0.0681. The model inputs are the spindle speed, feed rate, and statistical feature of acceleration and force signals measured during machining. Based on the model output, the GWO finds the optimal speed and feed of machining, which minimizes the tool wear. The online control of tool wear is achieved through a neural controller. The control strategy is simulated in MATLAB and validated in real time. The validation results show the efficiency of the adaptive control system with an improvement in tool life by 12% compared to conventional machining.
Important Links:
Go Back
