Design And Implementation Of Industrial Equipment Maintaince Platform Based On Hadoop

With the rapid development of China's industrial technology,a large number of automated production equipment has been put into use,thus greatly improving China's industrial level.However,as the functions of the equipment continue to improve,the links between the various parts are more closely linked.If any glitch is ignored,it may cause a chain reaction,causing greater damage to the equipment and even affecting the production process of the entire production line,bringing irreversible economic losses to the enterprise.Therefore,the equipment needs to be carefully maintained,so as to check the hidden problems of the equipment early and extend the service life of the equipment.Moreover,with the maturity of digitalization and networking technologies,the level of informationization of enterprises has been continuously improved,and a large amount of device status data has been collected through sensors and other devices.Using the massive historical data to guide the repair and maintenance of equipment,which is of great significance to the equipment.However,due to the heterogeneity,speed and complexity of industrial data,traditional data storage and analysis platforms are difficult to meet the actual requirements.Therefore,this thesis uses Hadoop cluster to realize an efficient industrial big data platform.The main work of this thesis is as follows:1)Based on the analysis of MapReduce distributed computing framework and distributed file system HDFS,the solution of Hadoop-based equipment maintenance platform is given,and Hadoop cluster is built.2)Researched the current mainstream big data storage and management solutions,applied the Mysql database and distributed file system to realize the storage solution of industrial data,and solved the problem of big data collection and storage;3)According to the actual needs,design and implement the equipment maintenance system,monitor the running status of the equipment in real time,and match and filter the specific rules of the data to complete the real-time monitoring and alarm;4)In order to improve the processing speed of MapReduce and the storage capacity of Hadoop cluster,this thesis takes the sensor monitoring data of the production equipment of an auto parts manufacturing enterprise as the research object,proposes a data compression strategy.At the same time,in order to further improve the compression performance,The compression strategy is optimized.