Industrial Production Process Based On Complex Network Theory Research On The Method Of Constant Dection

Since the 90s of last century,the application of condition monitoring technology in industrial production process and the detection of abnormal faults in monitoring parameters have become one of the important tasks of the factory.With the rapid development of social and economic development,the status of industrialization in the whole society is growing,and the contribution to society is increasing day by day.Among them,in the power gesneration process,the operation process is not many in artificial way,so a large number of enterprise funds were used for the upgrading of power generation equipment,more sophisticated control equipment,testing equipment and technology updates,relevant personnel only need to monitor the instrument,in order to achieve the real process of no state continues in normal operation,and once the whole process of any failure caused during the outage,will cause a lot of waste of resources and funds,and even lead to a regional chain of power,is harmful to the corporate image,this is the most reluctant to see.In recent years,although the related equipment and technology of the power plant are becoming more and more perfect,a large scale of power outages and power outages still occur.The combination of data analysis and condition monitoring technology has become the main trend of factory intelligent detection,safety and predictability.Relying on the industrial system of the existing physical topology,enterprise monitoring system to monitor the real-time data and the simulation data based on parameter correlation is established by using network theory and method of complex network system in science based on the calculation of the potential relationship between each node parameters to characterize the network structured index to the operation state of the system and the complex networks,determine the current whether the system of abnormal operation using the system level structure index,from the system level and the local level,revealing information transfer,cascading failure rule,then the system may produce abnormal condition anticipation.It is mainly divided into:(1)building complex network based on simulation data;(2)analyzing the relationship between network topology characteristics and system anomalies,and obtaining conclusions;(3)constructing complex network with real data(4)real data,verifying anomaly detection conclusion.