Design And Implementation Of Equipment Health Monitoring System Based On Internet Of Things And Big Data Analysis

With the development of traditional manufacturing industry,the variety of industrial equipment is becoming more and more diverse,and the number of equipment in the factory is increasing gradually,this leads to the fact that industrial managers are hard to grasp the health status of industrial equipment,and the management of equipment's health and the plan of replacement of equipment are always the focus of industrial management.At this time,the rapid and convenient detection and prediction of the health status of industrial equipment has become the key demand to be solved in the field of manufacturing industry.At present,the Internet of things technology,the big data technology and machine learning technology are increasingly mature,and the combination of the three with many fields has become the trend of development.The three have natural application advantages in the health management of industrial equipment.With the national intelligence manufacturing strategy direction put forward,the traditional manufacturing industry needs more new technologies to transform it.Therefore,the combination of three technologies is an important method to solve the detection and prediction of the health status of industrial equipment.Aiming at the demand for intelligent management in industrial manufacturing industry,this thesis proposes a solution for detection and prediction of the health status of industrial equipment with combining the above three technology.According to the factory's drilling equipment(radial drilling machine),the technology of Internet of things is used to solve the data collection of industrial equipment,and the storage problem of massive industrial data is solved by big data technology.Machine learning technology is used to detect and predict the health status of industrial equipment.This thesis designs and implements a set of industrial equipment health monitoring system which can be used in manufacturing industry.The usability and effectiveness of the software platform is verified through experiments and tests.First of all,the research status and development of wireless sensor network(WSN)and fault detection and health management(PHM)technology are briefly described in this thesis,and demand of equipment intelligent monitoring which the manufacturing industry facing is analyzed,and introduce related technologies of the industrial equipment and health monitoring system.Then,detailed analysis of the demands of data collection,big data storage platform,the industrial equipment judgement and data real-time query function.Based on demands analysis and related technologies,the whole system architecture is composed of three parts:the wireless sensor network layer,the big data platform layer and the front end data display layer.Then,this thesis introduces the deployment and construction of wireless sensor network,the effective and practical technology selection for deployment of big data platform,and the implementation of related machine learning models in this system.While,the machine learning model selects the optimal model by using a variety of model contrast methods.Based on these,the implementation of various functions are introduced in this thesis,such as real-time display of collected data,periodic query of device status data and real-time display of status data.Finally,in the production environment,the function and performance of each module in the industrial equipment health inspection system is tested in detail,which verifies the effectiveness and practicability of the design scheme,and in the part of the machine learning model,the model used in this thesis and the previous research model are compared.After testing,the system implemented in this thesis meets the needs of industrial equipment health monitoring,and this can solve the pain points of the health management of industrial equipment to a great extent.At the end of this thesis,a detailed summary of this thesis is proposed,the improvement of the related design scheme is analyzed.