基于全息时标量测数据挖掘的配电网设备健康状态诊断分析

大数据、数据挖掘等新技术的出现和进步，为建设智能配电网提供了新的技术手段。为实现对配电网运行状态的真实还原、精细分析和精准预测，详实有效的配电网设备运行数据记录和支撑是关键，研究了配电网全息时标量测数据的变化即存储技术，基于全息时标量测数据研究了配电网设备健康状态诊断的方法，对配电网历史数据以及模型信息等进行了深入的数据挖掘，通过聚类分析、线性回归算法、熵权法等建立了设备状态诊断模型和评价体系，实现了对设备故障评估和预警分析等，为及时发现配电网的薄弱环节，保障配电网设备的安全稳定运行提供了有效手段。系统已在地市供电公司的配电网诊断方法研究与实现项目中得到实际应用，很好地满足了地市供电公司的配电网精益化管理需求。

HEALTH DIAGNOSIS METHOD OF POWER DISTRIBUTION EQUIPMENT BASED ON HOLOGRAPHIC TIME-SCALAR MEASUREMENT DATA

The emergence and progress of new technology, big data, data mining, provide a new technical means for building intelligent distribution network. In order to realize the real reduction, fine analysis and accurate prediction of the distribution network operation state, the real, detailed and effective distribution network equipment running data recording and support is the key. The technology of storing all changed data of holographic time-scalar measurement data of the distribution network is studied. On the basis of data recording, health diagnosis method of power distribution equipment is studied and in-depth data mining on the distribution history data and model information is carried out. The equipment state diagnosis model and the evaluation system are established by cluster analysis, linear regression algorithm and entropy method, and the equipment fault assessment and early warning analysis are realized. The weak links in distribution network can be found in time to ensure the safe and stable operation of the distribution equipment. The system has been applied in the construction project of research and implementation of distribution network diagnosis method based on big data technology of district power supply company. It can well meet the demand of lean management needs of distribution network of district power supply company.