Process-oriented unstable state monitoring and strategy recommendation for burr suppression of weak rigid drilling system driven by digital twin

Robots have been widely used in machining due to their excellent expansibility and high flexibility. However, the robot is a weak rigid system, and its machining process is unstable. So, it is easy to produce burr due to chatter. A burr suppression pattern of the weak rigid drilling system driven by the digital twin is proposed to solve the above problem, including unstable state monitoring and strategy recommendation. The machining state is mapped in the information space through the hidden Markov model in the digital twin system. The quality of each machining feature can be tracked based on the model evolution rules. Based on the machining state mapping and quality tracking process, the machining state monitoring is realized. The drilling strategy storeroom is built based on the genetic algorithm, and the abnormal monitoring information is used to search for the optimal strategy. Then, the feedback strategies are recommended for improving the machining quality. Finally, the effectiveness of the proposed method is verified by a case of weak rigid drilling.