Research And Practice Of Steelmaking-continuous Casting System In The Model Library For The Iron And Steel Industry

The iron and steel industry is the basic pillar industry for the development of the national economy,and the iron and steel industry model is the foundation for supporting the intelligent manufacturing of the iron and steel industry.However,due to the strong professionalism of model development in the steel industry,model developers are accustomed to independent development,there is a lot of repetitive work,industry knowledge and data barriers make the model "island effect".Therefore,it is urgently needed for national-level research institutes and institutions of higher learning,based on the characteristics of the iron and steel industry,to integrate the iron and steel industry model development environment and developer communities,to accurately match the business needs of iron and steel enterprises with developers,and to integrate industrial mechanisms,advanced technology,Industrial technologies are encapsulated into industrial models,and the steel industry model library system is constructed and gradually extended to steel enterprises.This paper explores the support environment of the steel industry model,studies the method of constructing and managing the steel industry model library,and initially forms a model development and management standard,and integrates the developed steel industry model on the ModelArts platform,such as applying it to steel mills.The steelcontinuous casting system’s steelmaking-continuous casting scheduling model and steelmaking production planning model based on an improved genetic algorithm.(1)Research the model support environment of the iron and steel industry,conduct demand analysis from the perspective of model developers and model users,and determine the solution of using Huawei Cloud ModelArts platform as the model integrated development environment.On this basis,combined with ModelArts’ existing development environment and related knowledge of the steel industry,specific functional modules that need to be added are proposed.With the knowledge of the steel industry as the background,the developer community environmental management mechanisms such as user management mechanism,development process management mechanism,model management mechanism and developer training mechanism are introduced.(2)Based on the needs and process characteristics of the steel industry,participated in the research and design of the functional modules and databases of the steel industry model library.The model classification rules are analyzed,the knowledge graph is constructed by the method of graph database,and then the model search engine based on the knowledge graph is established.In the cross-platform data exchange of the model library,the Open API architecture is used to realize the openness of the model interface,and JSON is used as the data exchange format,and the test method of the model library is studied.Finally,a model development and classification management standard specification was initially formed.(3)Aiming at the steelmaking-continuous casting system of steel mills,this paper improves the basic genetic algorithm by proposing adaptive crossover operators and mutation operators,and applies it to the steelmaking-continuous casting scheduling problem.The casting scheduling problem was modeled,and the effectiveness of the improved algorithm was verified through simulation numerical experiments.At the same time,a steelmaking production planning model has been developed,and the above models have been applied to certain steel companies.Finally,the deployment of the model is introduced in detail.