A Mould Quotation System Of Blanking Die Based On Knowledge

Fast and reasonable mould quotation is always the goal that pursued by the mould manufacturer. Now the study of reasonable and accuracy mould quotation has obtained some achievements, but the fast mould quotation system is still required further research and development. In 2008, the global financial crisis has been involved in mold manufacturing, and the impact is deeper and deeper, so the 'rapid response' capability is more necessary to mould manufacturing.In this paper, the traditional cost estimation methods and their characteristics are analyzed in detail. According to the needs of enterprises, an approach of rapid die cost estimation is explored which makes the blanking die as research object. The detailed quotation of small-scale die is modified and improved. A die fast quotation model is constructed by standard mould base method and BP neural network. In addition, an inference engine of the production rule is also integrated. In the object-oriented environment, the open quotation system is established successfully.First of all, the blanking parts and mould parts features that effect the mould cost is classified and coded, and on the basis of this an inference engine of the production rule is developed. Thus blanking process scheme and mould type are determined. Then fills out the 'Questionnaire' window message and achieves automatic quotation. Detailed quotation or the quotation of standard mould base method is used for the miniature blanking die. Detailed quotation is a method that gets the accumulated mould prices according to mould template costs, standard parts costs, manufacturing costs and design costs. As the auxiliary part of the detailed quotation, standard mould base quotation estimates entire mould prices by the standard die base price searched. Cost estimation method based on BP neural network is used for medium scale and large scale blanking die. Three BP neural network model for die cost estimate is set up. Then the weight and threshold of BP neural network are optimized by genetic algorithm, so the problem that the network is easy to fall into local minimum is solved and the convergence rate is improved. Operation has proved that the system function met the design purpose.The development idea and mathematical model of blanking die and consideration of system development platform proposed in this paper have reference practical value for other kinds of mould quotation.