Optimal tuning of control gains for rigid tapping processes using a learning automata methodology

Rigid tapping is a novel tapping process and is extensively used to produce internal threads on a manufactured mechanical part because of its precise and efficient performances. However, it is difficult in practice to design control gains for controlling rigid tapping processes using systematic approaches because the operational environments in the rigid tapping processes are usually unknown and cannot be predicted or modeled precisely. Tuning a set of control gains that could achieve good tapping results in rigid tapping processes is therefore a great challenge for users. In this study, we investigate the use of the learning automata methodology in optimal tuning of the control gains of rigid tapping processes operated under unknown operational environments. The rigid tapping results achieved on a CNC tapping machine indicate that the developed learning automata tuning method (consists of two tuning phases) can effectively tune the optimal control gains. As compared to tapping results where the control gains are obtained from a manual, the developed tuning method provides optimal control gains so that the CNC tapping machine efficiently and precisely produces internal threads that pass thread gauge tests.