基于关联规则和竞争凝聚算法的高压电缆缺陷识别模型

为便于运维人员及时准确判断高压电缆缺陷类型,提出了一种基于关联规则和竞争凝聚算法的高压电缆缺陷识别模型。该方法基于现有在线监测、离线试验和运维系统等数据,利用关联规则挖掘出不同缺陷类型与状态数据的关联关系,并建立电缆缺陷关联规则库;针对现有关联规则算法只能处理布尔型数据局限,采用竞争凝聚算法对连续型数据进行离散化处理;最后通过实例对所提的模型进行分析验证,仿真结果表明该模型识别准确率高、求解效率高、解释性好,有助于运维人员对电缆进行及时动态的维护管理。

Defect Recognition Model of High-voltage Cable Based on Association Rules and Competitive Agglomeration Algorithm

In order to facilitate the operation and maintenance personnel to judge the defect types of high-voltage cable timely, this paper proposed a defect recognition model of high-voltage cable based on association rules and competitive agglomeration algorithm, which relies on the existing data of online monitoring, offline test and operation system. The association relation between different defect types and state data is found by using association rules, and a cable defect association rule library is established. Considering that existing association rule algorithm can only deal with Boolean data, the continuous data is discretized by competitive agglomeration algorithm. Finally, the model was analyzed and validated by a case, and the simulation results showed that the model has high recognition accuracy, high solving efficiency and good explanatory ability, which is helpful for operation and maintenance personnel to conduct timely and dynamic maintenance and management of cables.