A mold modular forming recommendation method and system

Abstract

The application provides a mold modular forming recommendation method and system, including: obtaining historical mold design schemes, extracting product features, process parameters and corresponding mold module combinations in each scheme, and constructing a historical scheme database; based on mold function unit decomposition, a module library including a basic mold frame module, a forming module, a core-pulling module and an auxiliary module is established; a rule knowledge base is constructed based on domain expert knowledge; a recommendation model is constructed based on a deep learning network, a multi-objective optimization loss function is constructed, and joint optimization is carried out in combination with the attributes of the module library and the constraints and guidance of the rule knowledge base, so that the recommendation model is trained; the characteristics and process parameters of a target product are input into the recommendation model, and a mold module combination scheme with the highest comprehensive score is output; the method trains the deep learning recommendation model through the multi-objective optimization loss function, balances the economy and timeliness while ensuring the technical feasibility, and improves the design efficiency and quality.

Images