基于气象信息的安徽电网故障预警研究

发布时间：2018-08-19 09:41

【摘要】：随着电网的快速发展和强对流天气的增多,由恶劣天气造成的输电线路故障频繁发生,严重威胁了电网的安全可靠运行。雷电现象生消时间短、变化快,引发的线路跳闸事件在电网故障统计中所占比例极大。大风作用下线路会发生偏斜,引起的风偏放电故障重合闸不易成功。各种自然灾害尤其是大范围的灾害性天气对输电线路造成的损坏比电气事故的影响更为恶劣严重,计及气象因素对电网故障进行预警研究具有重要意义。本文基于气象信息对安徽电网的故障进行了预警研究,主要包括由雷电和大风引起的故障,并基于气象信息给出一个电网静态安全风险评估体系的框架和模型。考虑到雷电现象生消时间短、变化快的特点,利用雷电在线监测数据,提出基于图像识别技术进行雷区识别,并基于三时次雷区信息进行雷电轨迹追踪和预测。同时,对安徽省落雷历史数据进行统计,得到杆塔的落雷概率。在此基础之上,结合输电线路雷击跳闸计算方法,利用雷区预测信息计算线路的雷击跳闸概率,实现线路雷害故障预警。结合历史数据进行算例分析,结果表明,基于三时次信息的雷区预测准确性较高,相应的线路雷击跳闸概率值也具有较高的准确率而可以作为雷击跳闸预警指标。分析风的构成,对脉动风风速进行模拟。利用几何方法分析悬垂绝缘子串的风偏响应,对输电线路在稳定风和脉动风作用下的风偏响应进行对比分析,得出脉动风对悬垂绝缘子串风偏的影响随基准风速的增大而增大的结论。从而以此为基础,提出利用g-h分布拟合脉动风风速而实现大风预警,并提出基于脉动风模拟利用概率统计方法计算输电线路风偏概率。结合历史数据进行算例分析,结果表明,g-h分布对脉动风风速的拟合效果较好,线路风偏故障概率预测也较为准确。将电网气象故障概率与电网运行状态相结合,提出基于D5000平台的电网静态安全风险评估体系,介绍了其框架与模型构成,给出基于气象信息的全网在线监测与评估的实现方案,为建设电网气象故障预警系统给出建议。
[Abstract]:With the rapid development of power grid and the increase of severe convective weather, transmission line faults caused by severe weather frequently occur, which seriously threaten the safe and reliable operation of power grid. The lightning phenomenon has the advantages of short generation time and fast change, which results in a great proportion of line tripping events in power network fault statistics. The fault reclosing caused by wind bias discharge is not easy to succeed because of the deflection of the line under the action of strong wind. All kinds of natural disasters, especially the disastrous weather in a large area, cause more serious damage to transmission lines than electrical accidents, so it is of great significance to study the early warning of power network faults taking into account meteorological factors. In this paper, based on meteorological information, the fault warning of Anhui power grid is studied, including the failure caused by lightning and gale. Based on the meteorological information, the framework and model of a static security risk assessment system for power grid are presented. Considering the characteristics of the lightning phenomenon, such as short duration and fast change, using the on-line monitoring data of lightning and lightning, this paper puts forward a method of minefield identification based on image recognition technology, and tracks and forecasts the lightning track based on the information of 03:00 minefield. At the same time, according to the historical data of falling mine in Anhui Province, the probability of falling out of pole tower is obtained. On this basis, combined with the lightning tripping calculation method of transmission line, the lightning tripping probability of the transmission line is calculated by using the prediction information of minefield, and the lightning damage fault early warning of the transmission line is realized. Combining with the historical data, the results show that the prediction accuracy of the minefield based on 03:00 information is high, and the corresponding lightning tripping probability value has higher accuracy and can be used as an early warning index for lightning tripping. The composition of the wind is analyzed and the wind speed of the pulsating wind is simulated. The wind deviation response of a suspended insulator string is analyzed by geometric method, and the wind deviation response of transmission line under the action of steady wind and pulsating wind is compared and analyzed. It is concluded that the influence of pulsating wind on the wind deviation of the suspended insulator string increases with the increase of the reference wind speed. On this basis, it is proposed to use g-h distribution to fit the wind speed of pulsating wind to realize wind warning, and to calculate the probability of wind deviation of transmission line based on the simulation of pulsating wind. Combining with the historical data, the results show that the distribution of g-h has a better fitting effect on the wind speed of pulsating wind, and the probability of fault prediction of wind deviation of the line is also more accurate. Combining the probability of meteorological failure of power network with the running state of power network, a static security risk assessment system based on D5000 platform is put forward, its framework and model structure are introduced, and the realization scheme of on-line monitoring and evaluation for the whole network based on meteorological information is given. Some suggestions are given for the construction of meteorological fault warning system in power grid.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM75;TM08

【参考文献】