Intelligent Recommendationsystem Of Aerospace Assemblymodel Base

In recent ten years,with the continuous improvement of China’s space science and technology hard power,the research field of space assembly model has also made long-term progress,but there are still some important technical problems to be solved in the application of the model.First of all,with more and more space model design and practical application,the information degree of space model management is low.Secondly,the lack of intelligence in the viewing and retrieval function of the large volume aerospace equipment model library not only reduces the use efficiency of the equipment model,but also seriously affects the breakthrough and innovation of aerospace assembly technology.Aiming at the informatization and intelligencization of space assembly model,the intelligent recommendation system with Java Web Web development technology,BM25 algorithm,SBERT algorithm and collaborative filtering recommendation algorithm,etc.is innovatively applied in the retrieval and recommendation function of space equipment model base.And realize the forum-oriented technical exchange.The system will effectively change the traditional model retrieval and management mode and realize the intelligent sharing of assembly model information.The intelligent recommendation system of the space assembly model base not only improves the efficient management of model information,but also facilitates the staff to consult the model,improves the work efficiency,and provides technical support for promoting the transformation and upgrading of space equipment technology to intelligence.The main research contents of this paper are as follows:(1)Firstly,the framework structure of intelligent recommendation system of aerospace assembly model library,key technologies of model retrieval and recommendation,and various algorithm models and program technologies required in the realization of other related functions are described.(2)Secondly,the feasibility of the system design is analyzed from three aspects of technology,operation and economy.According to the overall functional requirements of the system put forward by the use environment,the content is designed.At the same time,the information of the existing 2000 aerospace assembly models in the system is extracted and stored in the My SQL database as the core data source.(3)Thirdly,design specific solutions according to system requirements.According to the different functions,the system is divided into five functional modules:system management,intelligent search system,intelligent search system,model search and forum exchange.In order to improve the efficiency of system management,the introduction of permission management,the user is divided into three permission categories,and through the use of database E-R diagram to describe the connection between the database,complete the detailed design of each data table.(4)Finally,the business functions of each functional module of the system are described in detail according to MVC architecture,UML diagram and time sequence diagram,and BM25 algorithm and SBERT algorithm are compared and analyzed in terms of retrieval effect according to the actual effect drawing of aerospace assembly model library system,and the results of similarity calculation of three different users are compared through collaborative filtering.The most suitable algorithm is selected to verify the feasibility of the system.In terms of knowledge sharing,users can exchange technology through forums.