Abstract
The appropriate selection of dry hobbing control parameters can significantly improve cutting performances such as energy consumption and time. To overcome the problems of manually setting the value range of control parameters and numerical oscillation of the optimization results (i.e. low stability), a hybrid optimization approach is proposed for the improvement of hobbing performances under user evaluation using K-means clustering, multi-objective hunger games search (MOHGS) and TOPSIS. Firstly, K-means clustering is applied to obtain the cluster centers and case clusters based on the historical cutting cases. Considering the problem of control parameters to be optimized, the numerical range of control parameters is determined via the cluster centers and case clusters. Then, MOHGS is developed and applied to search the non-inferior control parameters and run for many times based on the multi-target optimization model of dry hobbing. Finally, all the non-inferior control parameters obtained are ranked by TOPSIS method based on the user evaluation of cutting performances, and the control parameter ranking first is for machining. Compared with other well-known approaches, it is proved that the proposed approach can effectively resolve the control parameter optimization problem of dry hobbing under user evaluation and has high stability.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call