Supervision Control for Quality Assurance in Milling Processes

Intelligent control algorithms are a key enabling technology for the smart factory concept of Industry 4.0. They must cope with complex, nonlinear and time-varying processes. Die and mould operations, in particular to milling, are known for their complexity; and improvements to process quality are of utmost concern, owing to economic factors, the drive to reduce lead times, and increasingly stringent customer requirements. Hence, a hybrid artificial intelligence (AI) controller (HaiC) is designed to supervise the spindle speed cutting profile of machining processes to ensure the desired surface quality of the machined part is achieved. It is comprised of artificial neural networks (ANNs) and fuzzy logic (FL) predictive models, and a proportional-derivative (PD) loop, to promote an efficient control based on the manufacturing requirements. Milling experiments were carried out for data collection for the development of the predictive models. The control part is developed and implemented in a simulation environment, and its performance is assessed through several machining scenarios. The results show that the controller provides high-accuracy spindle speed control for assuring surface quality in milling operations. Furthermore, it can also be used for offline optimisation of the spindle speed at the process planning phase.