A COMPARATIVE-STUDY OF NONLINEAR OPTIMIZATION AND TAGUCHI METHODS APPLIED TO THE INTELLIGENT CONTROL OF MANUFACTURING PROCESSES

The current investigation focused on neural-network-based control of m anufacturing processes utilizing an optimization scheme. In an earlier study, Demirci and Coulter introduced the utilization of neural netwo rks for the intelligent control of molding processes. In that study, a forward model neural network, employed with a search strategy based o n the factorial design of experiments method, was shown to successfull y control the flow progression during injection molding processes. Rec ently, Demirci et al. showed that the search mechanism based on the fa ctorial design of experiments method can be intolerable in time during on-line control of manufacturing processes, and suggested an inverse model neural network. This inverse model neural network was shown to b e beneficial as it totally eliminated time-consuming parameter searche s, but it required a harder mapping than the forward model neural netw ork and thus its performance was inferior. In the present study, the a uthors investigated two different optimization methods that were utili zed in making the search method of the forward control scheme more eff icient. The first method was Taguchi's method of parameter design, and the second method was a nonlinear optimization method known as Nelder and Mead's downhill simplex method. These two methods were separately utilized in creating an efficient search method to be used with the f orward model neural network. The performance of the resulting two cont rol methods was compared with each other as well as with that of the f orward control scheme utilizing a search strategy based on the factori al design of experiments method. Although the applications in this stu dy were on molding processes, the method can be applied to any manufac turing process for which a process model and an in-situ sensing scheme exists.