Research Of Web Expert System For A Transforming Method Of Mapping Natural Language To SysML Requirements Diagram Model

As the engineering systems developed become more and more complex,traditional systems engineering(TSE)becomes more and more difficult to deal with.At the same time,the information technology represented by modeling is also developing rapidly.Therefore,driven by demand traction and technology,model-based systems engineering(MBSE)comes into being.Model-based systems engineering has a great change in modeling language,modeling ideas and modeling tools.Compared with traditional system engineering,it has many irreplaceable advantages and it is a disruptive technology of system engineering.Product domain knowledge is essential in transforming the product requirements described by users in natural language into requirements models that drive product design,and in mapping requirements models to functional models.Therefore,it is of great significance to research and develop the transformation method of requirement description-requirement model-functional model mapping to achieve the positive demand capture of product conceptual design,eliminate design ambiguity and automate model driving.A transforming method of mapping natural language requirements texts to SysML requirements diagram model is proposed,and a Web-based software system is developed to verify the method.The work accomplished is:(1)A novel transformation algorithm converting natural language textual requirements description to formal SysML requirements diagram is proposed.After analyzing the imported requirements text,a word segmentation tool is employed to break up the text,and then the mapping relationship between SD model and ROM model is used to convert the SD model to ROM model,finally,the SysML requirement diagram model is generated by ROM model,the transformation between natural language and SysML requirement diagram was implemented.(2)A Web-based model transformation expert system is developed.The expert system is constructed from the three aspects of database,knowledge base and inference machine.Knowledge expression and inference algorithm are the focus of expert system.The binary tree and fuzzy rules are adopted to make the knowledge expression more complete.The control strategy of forward inference is used to design the inference machine,and combine the conversion from natural language to SysML requirement diagram to generate the SysML requirement diagram.(3)A Web-based expert system is designed.The traditional expert system is combined with internet technology.The expert system can be operated on the Web page.It is of great significance to solve the limitations of traditional expert system interaction.The expert system can be operated by the user on the Web page and obtain the desired results.(4)Through the requirement of water distiller,3D printers and stamping die shows this system from the import registration,log in to the demand,generation of each build model,and the process of generating SysML requirement diagram.The Web design based on mobile devices is added to make the expert system more adaptable to the development of the society and the trend of mobile office.