Prediction and geometric adaptive control of surface roughness in drilling process

Abstract

Surface roughness is an essential factor to evaluate the quality of component that decides the wear and fatigue properties and influences the quality of assembly. This research article focuses on real time control of surface finish in drilling using geometric adaptive control strategy. The dynamometer sensor and accelerometer are used to capture the force and vibration signals during drilling. A cubic SVM model is employed to model the surface roughness using the force, vibration and machining parameters. The accuracy of the prediction model is found to be 94 %, and the model is successfully used to control the roughness in drilling. The adaptive scheme uses a Neural Network (NN) controller to adjust the drilling parameters for ensuring the set roughness tolerance. The performance of the controller shows the potentiality of the presented methodology for the practical application in industries.

Keywords

• adaptive control
• cnc drilling
• roughness
• space vector machine (svm)