摘要
为了建设电能计量装置高效准确的智能诊断系统,提出了基于并行化朴素贝叶斯方法的电能计量装置故障智能诊断方法。针对电能计量装置大数据的特点,搭建电能计量装置大数据处理平台。根据计量装置具有的异常情况及故障类型,得到计量装置故障诊断处理流程。提出了Spark并行化朴素贝叶斯算法,用于在大数据平台实现计量装置故障类型的判别。实验测试及实验对比分析,验证了Spark并行化朴素贝叶斯算法在诊断计量装置故障类型上的可行性,准确性及高效性,尤其适用于海量数据环境下,说明智能故障诊断方法具有较强的应用价值。
In order to build an efficient and accurate intelligent diagnosis system for power metering devices,an intelligent fault diagnosis method for power metering devices based on parallel naive bayes method was proposed.According to the characteristics of big data of power metering device,a big data processing platform of power metering device is built.According to the abnormal situation and fault type of metering device,the fault diagnosis and treatment process of metering device is obtained.Spark parallelization naive bayes algorithm is proposed to realize fault type discrimination of metering device on big data platform.Through experimental testing and comparative analysis,the feasibility,accuracy and efficiency of Spark parallelization naive bayes algorithm in diagnosing fault types of metering devices are verified,which is especially applicable to the sea data environment.It shows that the intelligent fault diagnosis method proposed in this paper has a strong application value.
作者
李晖
陈清族
马汉斌
许梓明
Li Hui;Chen Qingzu;Ma Hanbin;Xu Ziming(State grid fujian electric power co.,LTD.,Fuzhou350003,China;State grid xintongyili technology co.,LTD.,Xiamen350003,China)
出处
《国外电子测量技术》
2019年第8期52-56,共5页
Foreign Electronic Measurement Technology
关键词
电能计量装置
故障诊断
朴素贝叶斯
electric energy metering device
fault diagnosis
naive bayes