| Complex industrial processes generally have the characteristics of long processes,complex working conditions,complex production process mechanisms,strong coupling between variables and large amounts of production indicators,and production indicators monitoring system plays an important role in the monitoring of complex industrial processes.It is an important support for efficient production.With the rapid development of cloud computing,big data and other ICT technologies,IT technology and OT technology are rapidly converging,and enterprises are transforming from digital to intelligent.As a result,enterprises have become more and more urgent to have intelligent production index monitoring systems.At present,complex industrial process indicator monitoring systems have obvious deficiencies in many aspects such as data insight,integration of experts and industry experience,real-time evaluation of monitoring results,system configurability,reconstruction and evolution,resulting in the existing indicator monitoring system is difficult to adapt to the requirements of intelligent development of enterprises.In addition,with the expansion of the scale of enterprises,the increasing complexity of production,and the widespread application of intelligent sensing technology,more and more data can be collected by enterprises,and it also brings many new challenges to the monitoring of production indicators.Therefore,this thesis uses configuration design ideas and a variety of technologies,such as data detection,data visual and visual analysis,software evolution to design and develop a configurable,reconfigurable,scalable and intelligent monitoring and analysis platform of production index with data exploration and analysis,integration of experts and industry experience,algorithm and model integration,real-time evaluation of monitoring results,and dynamic evolution functions.It can effectively enhance the intelligence of the monitoring system of production index for complex industrial process.Based on the project of enterprises-"the first phase of the transformation of the lorite powder ore suspended and magnetized roasting ore transformation project integrated automation MES project",according to the characteristics and monitoring requirements of complex industrial objects,the design and development of visual monitoring and analysis platform for production indicators of complex industrial process.And the platform was successfully applied to large-scale beneficiation plants in western China.This article mainly includes the following work:(1)Firstly,the research significance of this article is discussed.The current research status of visual monitoring in different industries and the research status of visual monitoring systems for industrial production indicators are analyzed.problem.Then,the existing problems of visual monitoring system of production indicator are pointed out.On this basis,combined with the characteristics of production indicators of complex industrial processes,the requirements of the production index visual monitoring and analysis platform,including functional requirements and performance needs,are analyzed.(2)According to the functional requirements and performance requirements of the platform,design each functional module of the platform.First,abstract and model complex industrial processes,and characterize a process by a seven-tuple.Based on this,a production process configuration design tool is designed.The indicator monitoring system for different production processes is constructed by configuration.The integration of expert knowledge and the fullconfiguration of the platform are realized,improving the readability and interactivity of the platformfor users.Secondly,it provided the design of open algorithm interface,can integrate the data analysis method,and lays the foundation for the analysis of the platform indicator data.Finally,a variety of visual analysis schemes are designed,and different visual schemes can be configured according to application needs,supporting visual analysis of indicator data detection and monitoring result evaluation,and providing a decision basis for users to dynamically adjust monitoring indicators or monitoring algorithms.(3)The platform uses a variety of advanced technologies to develop a visual monitoring and analysis platform for complex industrial process production indicators.The platform is based on the B/S architecture and adopts a front-end and back-end development approach;The front-end uses the Vue framework,the back-end uses the java SSM framework,and the front-end uses the RESTful API for data access;Based on Go.js library,the configuration of the configuration design environment is realized by Drag-and-Drop(DnD)technology;The factory design pattern is used to provide an open algorithm interface and support the integration of standard service interface algorithms.The system has integrated data analysis algorithms including random forest and LSTM;Use Bird’s eye view and Zoom-in/out for visual solutions such as multi-view monitoring to enhance insight into useful information extraction during monitoring;Enhance the user experience through the operation of human-computer interaction.On this basis,the development of six functional modules of the platform is completed.(4)In order to verify the effectiveness of the platform,taking Jiuquan Steel’s second-stage beneficiation plant as the specific application background,each functional module of the platform was verified with the actual production data on the site.In the end,the platform demonstrated its effectiveness from both a functional perspective and a performance perspective to meet the actual application requirements. |