摘要
针对传统基于无线传感器的配电网故障检测模型对大数据环境下的配电网故障数据存在诊断准确率低、故障诊断耗时较长以及经济效益较低的问题,设计基于大数据的配电网故障诊断预测模型,其采用RSIA模型对大规模故障信息进行智能搜索,计算出最优约简得到决策规则,实现对配电网故障发生位置的初步定位。采用基于模糊积分的故障诊断预测模型,根据初步诊断结果确定发生故障的候选元件及模糊测度值,根据拓扑信息以及元件的诊断结果形成不同相关联度的支持度集合,采用模糊积分融合技术确定模糊积分值构成故障可能性指标集合,根据该指标确定配电网故障发生的准确位置。实验结果说明,所设计模型能提高大规模配电网故障诊断的精度,缩短诊断用时,提高配电网的安全性。
The fault data of distribution network based on traditional wireless sensor fault detection model is based on large data environment,which is low in diagnostic accuracy,time-consuming in fault diagnosis and poor in economy. A distribution network fault diagnosis model based on large data was designed,which makes use of the intelligent searching ability of RS-IA for large-scale fault information and the optimal reduction of calculation decision rules to realize the initial location of distribution network fault. The fault diagnosis and prediction model based on fuzzy integral determines the candidate component faults and fuzzy measure values according to the preliminary diagnosis results,forms the related support set of different components according to the topological information and diagnosis results,determines the fuzzy integral value set of fault probability index by using the fuzzy integral fusion technology,determines the accurate distribution position of fault in the network according to the index.The experimental results show that the design model can improve the accuracy of the fault diagnosis of the large-scale distribution network,shorten the diagnosis time and improve the safety of the distribution network.
作者
程晓磊
王鹏
王渊
赵嘉冬
CHENG Xiaolei;WANG Peng;WANG Yuan;ZHAO Jiadong(Inner Mongolia Electric Power Economics and Technology Research Institute,Hohhot 010090,Nei Monggol,China)
出处
《电气传动》
2022年第2期61-66,共6页
Electric Drive
基金
内蒙古电力有限公司科研项目(510141190010)。
关键词
大数据
配电网
故障诊断
预测模型
模糊积分
预处理
large data
distribution network
fault diagnosis
prediction model
fuzzy integral
preprocessing
