基于气象因素的电网故障原因推理研究

发布时间：2018-08-29 17:16

【摘要】：电力系统故障发生的原因包括自然因素、电力设备本身因素和人为因素等,其中导致电力系统事故的主要自然因素为气象因素。我国领土南北跨度大,不同地区气象条件差异极大,给电力设备在不同气象条件下运行带来了巨大的风险。因此对电力系统故障的气象因素分析对电力系统故障排除及电力设备检修都有重要作用。在这个背景下,本文研究了面向多源异构数据的融合方法。研究符合安全防护规定的大容量数据跨安全分区融合技术,融合分别位于不同安全分区的若干系统的电网运行信息、设备状态信息、环境监测信息,并提出了基于多源数据的电网故障风险预警功能架构。本文提出了基于BP神经网络的电力系统故障气象原因分析方法,首先利用时序网络与模糊Petri网相结合的故障诊断方法诊断故障元件,再通过采集故障所在地区的历史气象因素相关数据和相应的电力设备故障率构建并训练神经网络,使得网络可以预测特定气象条件下的电力设备故障率,最后分析不同可能的气象条件下电力设备故障率,推测可能导致该元件故障的气象因素。本文针对江苏电力系统实际情况,侧重电网运行信息、设备状态信息、环境监测信息深度融合与综合应用,开展了基于多源数据融合的电网故障辅助分析系统的研制。该系统利用了多源数据来分析电网故障诊断的原因,大幅提高了故障分析和处理的效率,对于系统事故后快速恢复供电以及维持系统安全稳定运行具有重要意义。
[Abstract]:The causes of power system faults include natural factors, power equipment itself and human factors, among which the main natural factors leading to power system accidents are meteorological factors. The long span of north and south of our country and the great difference of meteorological conditions in different regions bring great risks to the operation of electric power equipment under different meteorological conditions. Therefore, the analysis of meteorological factors of power system faults plays an important role in the troubleshooting of power system and the maintenance of power equipment. Under this background, this paper studies the fusion method for multi-source heterogeneous data. This paper studies the technology of large capacity data fusion across security zones in accordance with the safety protection regulations. It integrates the grid operation information, equipment status information and environment monitoring information of several systems located in different security zones. The function framework of power network fault risk warning based on multi-source data is put forward. In this paper, a fault meteorological cause analysis method based on BP neural network is proposed. Firstly, fault diagnosis method based on time series network and fuzzy Petri network is used to diagnose fault components. Then the neural network is constructed and trained by collecting relevant data of historical meteorological factors and the corresponding failure rate of power equipment, so that the network can predict the failure rate of power equipment under specific weather conditions. Finally, the failure rate of power equipment under different possible meteorological conditions is analyzed, and the meteorological factors that may lead to the failure of this component are inferred. Aiming at the actual situation of Jiangsu electric power system, this paper focuses on the deep fusion and comprehensive application of power network operation information, equipment status information and environmental monitoring information, and develops a power network fault aided analysis system based on multi-source data fusion. The system uses multi-source data to analyze the causes of power network fault diagnosis and greatly improves the efficiency of fault analysis and processing. It is of great significance for the rapid recovery of power supply after system accidents and the maintenance of safe and stable operation of the system.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM711

【相似文献】