Data Mining Technology In The Fault Diagnosis Of Electrical Equipment In Mill

Developing with the modern science and technology promptness, the Iron and Steel processing Industry is developing rapidly toward Large-scale-rization, complication, systematization and automation. The Development brings forward requests to the safe reliability of electrical equipment of mill more strictly. Because of long-term operation, there're lots of hidden dangers in electrical equipment of mill. Huge losses will be made by the fault of it. Therefore, electrical equipment of mill of fault diagnose is regarded more important by people.First, based on the in-depth study on the theory of fault diagnosis of electrical equipment of mill, the main electrical machine--which is the main electrical equipment for driving the rotation of mill—was taken as the object of our diagnosis. The configuration of the machine, the types of fault and some commonly used fault monitoring technology were researched and some problems existing in the on-line monitoring system of the bar material production line were analyzed. Then the idea of introducing the OLAM mining technology, which is based on the association rules, into the field of fault diagnosis of electrical equipment in mill was be proposed creatively, in order to mine out the active and instructional information lied behind the monitoring data for improving equipment security and economic benefits.Second, warehouse technology, OLAP technology, data-mining technology and OLAM technology were researched and based on the in depth study on theory and algorithm of association rules mining, the feasible association rule mining algorithm was selected, so that it can be combined with data cube to form an OLAM model, which is based on association rule. Then, studied the configuration of the model combined with the fault diagnosis of electrical machine.Finally, because of large amounts of motor data stored in monitoring system, take Iron & Steel Company for example, under the environment of the Windows XP operating system, a data mining model of association rules is developed by using SQL Server 2005. We used the vibration test-bed's fault simulation data to train and check the model, the results indicated that the rules which is produced by association rules model can offer decision for the fault diagnosis.