Research Of Elevator Big Data Security Mining Platform Based On Hadoop

In recent years,with the rapid growth of elevator usage in the country,and the lack of elevator maintenance and supervision,elevator safety accidents have occurred,which seriously affects the personal safety of residents,so the safety of elevators in recent years It has aroused widespread concern in society.The increase in the use of elevators has led to a sharp increase in the data in the elevator supervision system database.Therefore,how to extract and analyze the daily operation of various brands of elevators in various places from the massive data,and plan out the targeted elevator safety maintenance and maintenance measures,thus reducing the occurrence of elevator accidents is particularly important.Permanent magnet synchronous motor(PMSM)without position sensor technology is becoming more and more widely in the field of motor control in application,for example in the application of new energy electric vehicles has become a development trend.The main research content of this subject is based on Hadoop's big data platform to mine and analyze the massive elevator data in the background of elevator remote monitoring system.Based on Hadoop's smallest cluster,a data mining and analysis platform based on Hadoop is designed.The platform is mainly composed of data.The conduction module,the data preprocessing module,the data mining module and the scheduling module are composed of four modules.The data transmission module mainly implements data transmission between the relational data block SQL Sever and the HDFS component in the hadoop;the data preprocessing module mainly performs operations such as cleaning,integration,conversion,protocol,etc.on the massive elevator data,so as to facilitate better The data mining module is mainly for excavating the elevator data after data preprocessing.In this module,the cluster analysis algorithm K-Means and association rule algorithm Apriori is improved and the Hadoop MapReduce computing framework is used.Parallelization is realized.Finally,all the modules are integrated by the scheduling module,and a data mining and analysis platform based on Hadoop minimum cluster is realized.1.This paper suggests that the types of permanent magnet synchronous motor and the relevant mathematical model.Some hypothesis conditions,permanent magnet synchronous motor specific mathematical expressions are presented.In this paper,the data in the database of elevator remote monitoring system is used as the data source of analysis and mining.The improved clustering association algorithm is parallelized in Hadoop's MapReduce computing framework,which improves the scalability of data mining and analysis platform and accelerates data processing.speed.The improved twoalgorithms are used to cluster and analyze the elevator data of several communities in the past three years.(1)Using the cluster analysis algorithm to analyze the elevator failure situation of each residential area and each elevator brand,and carry out key maintenance and maintenance on the elevator with high failure.(2)Using cluster analysis algorithm to analyze the fault repair rate of each maintenance company.(3)Using the association rule mining algorithm to analyze the specific frequent accidents of the community and the elevator brand,combined with the results of the cluster analysis,perform targeted maintenance and overhaul work,thereby reducing the occurrence of elevator accidents.