Energy analytics and machine learning modeling in milling of difficult-to-machine alloys
Manufacturing consumes approximately 20% of total energy consumption in the U.S. Energy consumption in machining, an important manufacturing cluster, is substantial, contributes to a large fraction of manufacturing cost, and has a significant environmental impact. On the other hand, energy consumption at the process level is directly related to surface integrity of machined part. Therefore, it is critical to investigate energy consumption in machining for process sustainability and surface integrity. This dissertation consists of seven chapters: (1) The first chapter gave a comprehensive literature review on energy consumption in machining; (2) The second chapter compared the characteristics of energy consumption in dry milling vs. flood milling of Inconel 718 superalloy at machine, spindle, and process levels; (3) The third chapter has developed an innovative holistic concept of cumulative energy demand to evaluate the energy demand in milling Inconel 718 from the life cycle perspective; (4) The forth chapter has proposed a hybrid machine learning model by integrating cutting mechanics to predict net cutting specific energy; (5) The fifth chapter has coined a new concept of energy-based process signature to establish the relationship between net cutting specific energy and surface integrity; (6) The sixth chapter has proposed a real-time energy consumption monitoring framework at machine and enterprise levels; and (7) The last chapter has summarized the conclusions and proposed the future work.