Research On Perception, Fusion And Visualization Of Digital Workshop Data

The “Intelligent Manufacturing Engineering” was launched as the five key projects,aiming to promote the transformation of the manufacturing industry to intelligence successfully.The construction of digital workshop is the crucial part for enterprises to realize intelligent manufacturing.The independent decision-making,self-assembly producing,equipment interconnection,automatic sensing,data fusion,real-time analysis and visualization of the workshop have got widespread attention in the academic and industrial circles.In the context of intelligent manufacturing and industrial big data,we take digital workshop equipment as our research object,and focus on the sensing methods of digital workshop manufacturing equipment data,data fusion and analysis technology,digital workshop visualization technology,to present data fusion and analysis results.Real-time monitoring and evaluation of digital workshop equipment,and construction of real-time operation status monitoring and evaluation platform,which provide assurance for digital workshop equipment safety,stability,high efficiency and green production.The main contents and innovations are as follows:(1)Aiming at the problems of data perception and aggregation,integration and analysis,sharing and application of digital workshop equipment,the research focus on the "poly,communication,and use" of equipment data,then the digital workshop data perception,fusion and visualization technology scheme and system architecture are proposed.And we introduce cloud computing,big data,Internet of Things,artificial intelligence and other technologies to solve bottlenecks,including large data storage,fast acquisition speed and complex structures in digital workshops.(2)In view of the problems that data perception faced,which include the inconsistent industrial protocol standards,the poorness in data openness and the difficulties in protocol adaptation,protocol resolution,and data interconnection,the research introduces the OPC UA architecture to solve heterogeneous data,heterogeneous network interconnection and other issues.Then we research the OPC UA architecture as the basic standard for data interconnection,and design a digital workshop equipment data sensing system based on OPC UA service equipment data access interface,to facilitate access to workshop equipment data in heterogeneous system networks such as SCADA,MES,ERP,and to solve the problem of heterogeneous device data and heterogeneous network interconnection(3)For data fusion and analysis,it is difficult to obtain enough fault samples to train the algorithm model to ensure its robustness and generalization ability in the application of equipment fault diagnosis.Therefore,a fault diagnosis data fusion analysis method based on few-shot learning is proposed.This method not only effectively improves the accuracy of the fusion analysis algorithm model in small data sets or unbalanced data sets,but also ensures the robustness and generalization ability.Few-shot learning methods are also applicable to other data fusion and analysis applications.In addition,in order to facilitate scholars to apply few-shot learning to the fusion and analysis of other equipment data services,we have disclosed the small sample fault diagnosis algorithm model and code.(4)Aiming at the problems in diagnosing multiple operating conditions and multiple faults caused by the difficulty in obtaining high-quality and massive operational fault data,a fault diagnosis model based on deep transfer learning is proposed and used in the equipment health assessment model,which uses existing knowledge learned from a large number of related different working conditions or different types of fault data to assist in the learning of a small number of new working conditions or kinds of new knowledge as quickly as possible,making full use of relevant equipment to collect operational fault data,effectively alleviating the difficulty of obtaining high-quality and massive operational fault data,and promoting the fault diagnosis and remaining life prediction of the equipment better and faster.It can also effectively solve the problem of data fusion and life prediction of equipment with multiple working conditions and multiple faults.The fusion analysis method of transfer learning is also applicable to other workshop equipment data service fusion and analysis applications.(5)Based on basic theoretical research,a digital workshop data sensing,fusion and visualization system was developed.At present,the system has been put into use in three aerospace enterprises and can operate stably,which fully utilizes and mines digital workshop equipment data to decrease equipment failure rate,extends the service life of the equipment,reduces the production cost of enterprise,improves the production efficiency of enterprise,and provides guarantee for the enterprise to realize the safety,stability,high efficiency and green production of the digital workshop.Therefore,this study has important theoretical and practical significance.