File(s) under permanent embargo
Performance estimation in drilling using artificial neural networks
thesisposted on 2023-05-26, 18:29 authored by Moore, Timothy John
Drilling is often carried out as one of the last steps in the manufacturing production of a part and demands process reliability. The work piece would have undergone extensive machining before and thus the final drilling demands considerable attention. It is apparent that any optimisation made to the drilling process will make manufacturing more productive and improve quality. The estimation of the process outcomes is necessary for reliable optimisation techniques. Traditional methods of performance estimation develop mathematical models and relationships, for individual performance estimation, for a set of process parameters. With advances in on-line control of machine tools there is a need to predict more than one performance feature. Artificial intelligence offers an alternative method for performance estimation and one that can perform simultaneous estimation of more then one performance feature. This project aims to use artificial intelligence, more specifically artificial neural networks (ANN), for the simultaneous prediction of the performance measures of thrust and torque in the conventional drilling process. An initial investigation into the relationship between hole oversize and the vibrations in two planes will also be undertaken. This work is seen as not only a step towards establishing intelligent tools for machining performance estimation, while addressing the mathematical and scientific basis of machining science, but also as a step towards the use of artificial intelligence for on-line control of the conventional drilling process.
Rights statementCopyright 2000 the author - The University is continuing to endeavour to trace the copyright owner(s) and in the meantime this item has been reproduced here in good faith. We would be pleased to hear from the copyright owner(s). Thesis (M.Eng.Sc.)--University of Tasmania, 2001. Includes bibliographical references