【摘要】 在解决电网配网运行过程产生的故障诊断问题时，需要考虑到当前大电网运行过程中产生数据库的规模大、维度高、模态多样及类型复杂等特点，而故障诊断中应用传统关联规则挖掘算法已无法适应处理大数据库数据量的实际需求。与此同时，以人工鱼群算法为代表的新兴智能算法在实际社会生产中得到了广泛的应用，因而提出一种基于人工鱼群算法改进Apriori数据挖掘算法的配电网故障诊断方法，构建出一种新型的改进Apriori算法模型。结合某一实际配网运行产生的数据，通过矩阵实验室(matrix laboratory, MATLAB)仿真软件进行仿真验证所提到的新型算法，并与传统数据挖掘算法进行比较。实验结果表明，改进后算法的准确率较高，响应速度快，更适用于对大规模数据库的处理。更多还原

【Abstract】 When solving the problem of fault diagnosis during the operation of power distribution network, need to consider the characteristics of large-scale, high dimension, diverse modes and complex types of databases generated in the current large-scale power grid operation process, the application of traditional association rule mining algorithm in fault diagnosis has been unable to meet the actual needs of processing large database data. At the same time, the emerging intelligent algorithm represented by the artificial fish swarm algorithm has been widely used in the actual social production. Therefore, in this paper we propose a distribution network fault diagnosis method based on the improved Apriori data mining algorithm of the artificial fish swarm algorithm, construct a new improved Apriori algorithm model, and combine the data generated by an actual distribution network operation to verify the new algorithm through matrix laboratory(MATLAB) simulation software. Compared with the traditional data mining algorithm, the experimental results show that the improved algorithm has higher accuracy, faster response speed, and is more suitable for large-scale database processing.更多还原