基于关联规则挖掘的电网设备差异化状态检测

Differential state detection of grid devices based on association rule mining

WFP电网设备差异化状态检测方法在电网设备数量增加但数据量不变时加速比较低,因此设计一种基于关联规则挖掘的电网设备差异化状态检测方法。根据电网设备的实际参数,分析电网危害程度,简化风险评估分级,建立设备差异化状态风险矩阵;利用决策树模型挖掘得到设备历史运行数据中满足支持度要求的项集,并对决策树模型进行剪枝处理;最后得到数据之间潜在关联,完成电网设备差异化状态的检测。实验结果表明,在告警量较少的情况下,设计方法更加接近理想加速比,验证了方法具有较好的性能。

WFP’s differential state detection method for power grid equipment has low acceleration when the number of power grid equipment increases while the amount of data remains unchanged.Therefore,a differential state detection method for power grid equipment based on association rule mining is designed.According to the actual parameter of the power grid equipment,power grid damage degree is analyzed,risk assessment classification is simplified,and an equipment differential state risk matrix is set up.After that,the decision tree model is used to mine item sets that could satisfy the requirement of support in equipment historical operating data,and the model is also pruned.Finally,the potential connection between the resulting data is obtained to complete the detection of the grid differential status of the equipment.Experiment results show that the method is more close to the ideal acceleration ratio when the number of alarms is small,which verifies that the method has good performance.