Through the practice of developing an expert system FBCDES(Feed Box Conceptual De- sign Expert System)for the planomillingmachine in CIMS ERC(CIMS Engineering Research Center) in China,the paper intends to introduce a...Through the practice of developing an expert system FBCDES(Feed Box Conceptual De- sign Expert System)for the planomillingmachine in CIMS ERC(CIMS Engineering Research Center) in China,the paper intends to introduce a common problem-solving strategy for the mechanical product conceptual design expert system.Meanwhile some problems such as system architecture,knowledge representations and inference engine sets are also described.A yet more detailed discussion is given to the technical problems for realizing two key modules,that is concept design module and layout module.展开更多
A computing model employing the immune and genetic algorithm (IGA) for the optimization of part design is presented. This model operates on a population of points in search space simultaneously, not on just one point....A computing model employing the immune and genetic algorithm (IGA) for the optimization of part design is presented. This model operates on a population of points in search space simultaneously, not on just one point. It uses the objective function itself, not derivative or any other additional information and guarantees the fast convergence toward the global optimum. This method avoids some weak points in genetic algorithm, such as inefficient to some local searching problems and its convergence is too early. Based on this model, an optimal design support system (IGBODS) is developed.IGBODS has been used in practice and the result shows that this model has great advantage than traditional one and promises good application in optimal design.展开更多
文摘Through the practice of developing an expert system FBCDES(Feed Box Conceptual De- sign Expert System)for the planomillingmachine in CIMS ERC(CIMS Engineering Research Center) in China,the paper intends to introduce a common problem-solving strategy for the mechanical product conceptual design expert system.Meanwhile some problems such as system architecture,knowledge representations and inference engine sets are also described.A yet more detailed discussion is given to the technical problems for realizing two key modules,that is concept design module and layout module.
文摘A computing model employing the immune and genetic algorithm (IGA) for the optimization of part design is presented. This model operates on a population of points in search space simultaneously, not on just one point. It uses the objective function itself, not derivative or any other additional information and guarantees the fast convergence toward the global optimum. This method avoids some weak points in genetic algorithm, such as inefficient to some local searching problems and its convergence is too early. Based on this model, an optimal design support system (IGBODS) is developed.IGBODS has been used in practice and the result shows that this model has great advantage than traditional one and promises good application in optimal design.
