Construction and application of process knowledge graph based on deep learning
In response to the issues of high dispersion,weak structural integrity,and poor reusability of existing part process knowledge,a method was proposed to construct a process knowledge graph based on deep learning technology for process knowledge reuse.Firstly,the structure of process knowledge was analyzed and a model layer of the process knowledge graph was established.Secondly,a deep learning knowledge extraction algorithm was built,the pattern layer was used as the data pattern to guide the process knowledge extraction,and the process knowledge map data layer was established.Then,a process knowledge inference model was constructed based on the graph neural network deep learning algorithm,serving as the foundation for process recommendations.Finally,extensive experiments were conducted to validate the proposed method.The part process knowledge graph visualization system was implemented,providing an intuitive interface for users to retrieve and recommend process knowledge effectively.The planetary gear parts were chosen as a representative example for the validation of the process knowledge retrieval and recommendation functionalities.The research results demonstrate that the process knowledge identification accuracy of the method reaches over 80%,and the process recommendations accuracy exceeds 70%.Comparing with the previous models and traditional methods,the proposed approach shows significant improvements in automating the construction of process knowledge graphs and enhancing process reuse capabilities.This research contributes to the advancement of manufacturing technologies by providing a more efficient and reliable solution for managing and reusing process knowledge.The findings of this study not only validate the effectiveness and feasibility of the proposed method but also pave the way for future research in the field of process knowledge management and utilization.
process knowledge graph constructiondeep learning technologyprocess reuseknowledge extractionknowledge reasoning modelgraph neural networkmode layer and data layer
万方数据
维普
