AI Approaches for Cutting Tool Diagnosis in Machining Processes

R. Morales-Menéndez, A. Vallejo, L.E. Garza, F. Cantú (Mexico), and J.V. Abellán Nebot (Spain)


Diagnosis, Pattern recognition, Bayesian networks, Hidden Markov models, Machining Processes.


Monitoring of cutting tool systems are very important in machine tools and manufacturing equipment due to the im pact they have in quality products and economy produc tion. The cutting tool condition can be determined by di rect or indirect sensing methods. Indirect methods are the only practical approach that offers better results by exploit ing data sensor fusion techniques, which help to make a more robust and stable diagnosis. Different successful ap proaches from the Artificial Intelligence (AI) community are reviewed. A discussion of the implementation and eval uation of two AI techniques is done. Hidden Markov Model (HMM) based and Bayesian Networks based into an indus trial machining center are tested. Excellent results demon strated that HMM-based approach has a potential industrial application.

Important Links:

Go Back