Research On Key Technologies Of Internet Of Manufacturing Things Services For Small And Medium-sized Enterprises

With the rapid development of information technology represented by big data,IOT and artificial intelligence,a new round of changes has taken place in the global manufacturing industry.Cloud manufacturing,IOMT(Internet of manufacturing things)and other new generation of intelligent manufacturing models are emerging,which integrate information technology and manufacturing technology.At the same time,a new round of industrial revolution and increasingly fierce global market competition have brought severe challenges to the survival and development of the majority of small and medium-sized enterprises.In view of how to help small and medium-sized enterprises in the current market environment to obtain greater space for development,new intelligent manufacturing mode of IOMT is a good way.It is of great theoretical value and practical significance to build a public platform of IOMT service for small and medium-sized enterprises to help them realize networked manufacturing more quickly and thus improve their competitiveness.The research focuses on the architecture and key technologies of IOMT service platform for small and medium-sized enterprises,mainly including the following aspects:(1)The application mode and architecture of IOMT service platform for small and medium-sized enterprises are studied.Based on the analysis of the problems faced by small and medium-sized enterprises,the demand and characteristics of IOMT in small and medium-sized enterprises are summarized.On this basis,the realization idea of IOMT service system for small and medium-sized enterprises is represented,and the IOMT service platform is proposed.The application mode of the platform based on dynamic alliance and the application process of on-demand are analyzed from the perspective of life cycle management.Finally,the key technologies to realize the IOMT service platform are analyzed.(2)The description and matching of IOMT services based on semantics and reasoning are studied.The manufacturing resources in IOMT environment are classified and described,and the service description method of manufacturing resources is studied.On this basis,the description information of manufacturing services is subdivided,and the similarity matching algorithm of various information is established to help users quickly and effectively complete the search process in massive manufacturing services to obtain manufacturing services that meet the needs of users.(3)The optimization of IOMT service based on mixed information axiom under uncertain service quality is studied.Aiming at selecting the optimal service from the candidate set of matching manufacturing services,the service quality is taken as the measure index.According to axiomatic information principle,a service quality measurement tool called information entropy is given,and a hybrid uncertainty optimization method is proposed,which represents the system range by random variables and the design range by fuzzy variables.For the cost type,benefit type and fixed type indexes respectively,fuzzy simulation method is adopted to calculate the information amount,and the integration of the upper bound and lower bound as fuzzy number is converted into the weighted sum of discrete point integral.The optimal scheme can be obtained by comparing the information amount.(4)The scheduling of IOMT services based on I-NSGA-II is studied.By analyzing the scheduling process of IOMT services,the scheduling model of IOMT services is established,which includes multiple objective functions and constraints with respects of maximum completion time,total cost and total quality of service.An algorithm for scheduling model is proposed.For further solution,double chain coding(sub task sequence chain and service selection chain),fast non-dominant sorting method,binary tournament selection method,improved priority operation crossover and inversion mutation of sub task sequence chain are mainly studied as well as improved elite strategy.Detailed steps of the algorithm are given.Finally,an example is given to verify the applicability of the model and the effectiveness of the algorithm.(5)The prediction method of service performance of IOMT is studied.The performance model of IOMT service is established and a time serial-based performance prediction method for IOMT service is proposed.Firstly,the performance index system of IOMT service is established based on the characteristics of IOMT environment and the performance model of web service.Secondly,a time serial-based manufacturing service performance model is proposed to predict the manufacturing service performance of IOMT.Then the time series prediction method based on ARIAM model and LSTM are proposed according to different characteristics of IOMT service performance indexes.Finally,the feasibility and effectiveness of the proposed method are verified by simulation comparison of examples.(6)The prototype system is developed.Based on the above research,the prototype system of IOMT service platform for small and medium-sized enterprises is built.The functional model of the platform is proposed,and the prototype system is developed by using integrated development environment and tools.The principles and functions of each module of the prototype system are demonstrated by an example.