Abstract: With historical equipment database,weighted association rule algorithm is adopted to conduct data mining.By using weighted association rules,a model base is established.Based on the model base,fault diagnosis can be performed by using equipment monitor data.In the meanwhile,selforganizing competitive neural network model is used to determine the weight of hydraulic equipment in a steel enterprise.Three properties are considered.They are degree of importance,degree of vulnerability,and level of fault.The model takes these three properties of the fault information as inputs and determines the connection weights of equipment fault through sample training.Experiments show that this algorithm can improve the accuracy of fault diagnosis of hydraulic equipment.  

Key words: data mining, weighted association rules, neural network, fault diagnosis