| Smart factories are the urgent need and research hotspot for the transformation and upgrading of traditional manufacturing industries.With the guidance of the government,increasing number of traditional manufacturing companies begin to purchase robots to develop automated production lines and build smart workshops and smart factories for high efficient and unmanned production processes.However,monitoring of the equipment status and process quality of the production line is relatively weak,and the level of quality management and production operation is generally low.With the development of Internet of Things sensing and network control technology,it is possible to monitor and manage the production environment,equipment status and crafting process of smart factories in full parameters with a variety of sensors.Based on the packaging production demonstration line(smart factory prototype)of a research institute in Nanjing,a distributed data collection network was built with multiple sensors,and a large amount of data generated during the production process were collected and uploaded to the server in real time.Then through data analysis,the operation status of the packaging production line was real-time monitored,based on which the platform provided preventive warning information and maintenance strategies to ensure the smooth running of the production line,reduce operation and management costs and then maximize the benefits of the smart factory.The main work of the thesis includes:1.Overall design.According to the current status of the smart factory,the overall demands were analyzed and the information model of the packaging production demonstration line was constructed.In order to meet the needs of information integration and operation status supervision,the overall system design of system architecture and communication schemes was constructed.2.Operation and maintenance methods.An operation and maintenance scheme was proposed which integrates equipment health monitoring,early warning of production line status as well as operation and maintenance decision support.With the usage of deep learning algorithms,the relationship model of equipment in-service status,environmental conditions and life expectancy was designed and built,by which the health status of production equipment was evaluated.Moreover,a comprehensive early warning model for intelligent production lines was established with equipment as the core,providing reasonable strategies for efficient operation and maintenance.3.Data acquisition method.According to the overall scheme and operation plan,monitoring demand analysis was conducted,and data acquisition strategies were correspondingly designed.With sensors,communication gateways and other related hardware selected,distributed data acquisition network was constructed,and then the efficient lightweight data communication protocols were designed to meet real-time and reliability requirements.4.Platform construction and software development.The overall construction of the platform was completed.And with regard to the requirements of data analysis and data storage,the software architecture and server was designed.Then,the software of the integrated management platform was developed realizing efficient network communication between the field collection terminal,the client and the server.5.Platform tests.Functional and non-functional tests on the integrated management platform were performed.Functional tests included data acquisition,transmission and analysis.The content of the non-functional test is to test the throughput of the server to verify the stability and reliability of the platform.In this thesis,the Internet of Things sensing technology and data analysis algorithms were applied to the production process monitoring of the packaging production demonstration line.Preventive maintenance methods and operation and management strategies were proposed,and a comprehensive management platform was developed.The work is of great value to the construction and management of smart factories. |
PDF Full Text Request