【作者】 麻赛军；

【导师】 段其昌；

【作者基本信息】 重庆大学 ， 控制理论与控制工程， 2010， 硕士

【摘要】 输电线路是电能传输的重要电力设施,由导线、绝缘子、杆塔、线路金具等构成,其安全稳定运行不仅关系到国民经济发展,而且维系着社会安全稳定。然而输电线路常年暴露于野外,容易遭受自然因素的破坏,其中导线覆冰是影响输电线路安全的主要因素之一,覆冰引起的导线过载和导线舞动可能导致相间闪络、金具损坏、跳闸停电、倒杆(塔)、导线折断等事故,严重威胁电网安全运行。输电线路早期的覆冰发现和覆冰预测变得十分重要。覆冰监测系统主要由覆冰监测装置和上层管理系统两部分组成。本文对输电线路覆冰远程监测系统进行了总体方案设计,研究了以GPRS等无线网络为基础的系统总体构架,并对数据传输方式和软件功能模块做了详细设计。重点对覆冰上层管理系统覆冰信息处理方法进行了分析与研究,以采集到的覆冰图片和气象信息为研究对象,研究了覆冰预测模型和覆冰频率分析方法,提出了一种基于图像分割技术的导线覆冰厚度计算方法,采用改进的基于混合能量泛函模型的水平集算法分割覆冰图像边缘轮廓,并计算覆冰图像增加的像素面积,通过比对导线固有像素面积和导线直径,最终换算得到覆冰厚度和等效的覆冰荷载。通过计算一定时间内覆冰厚度的差值得出覆冰增长的趋势与速率,综合判断覆冰现状与严重程度,在灾害发生前做出预警,并对覆冰发展趋势进行预测。本文最后研究了覆冰监测上层管理系统的具体实现方法,包括覆冰数据库设计和功能模块的实现。并以历年厚度极大值为基础,实现了覆冰频率分析,得到覆冰厚度设计值。以采集到覆冰图片和气象信息为基础,实现了基于覆冰预测模型的厚度计算,重点对基于图像分割技术的覆冰厚度计算方法进行了仿真与计算分析,通过与实际测量数据的对比得出,本文图像分割算法能准确、快速的提取覆冰边缘轮廓,覆冰厚度计算结果接近实际值,能正确反映实际覆冰状况。更多还原

【Abstract】 Transmission lines are important power transmission facilities which are composed of conductor, insulator, fittings and tower, the safe and stable operation of transmission lines is related to national economic development as well as the security and stability of society. However, they exposure to the wild all year, and can be easily destroyed by the natural factors, where conductor icing is one of the main factors affecting the security of the transmission line, The overload and galloping of the conductor caused by the icing can lead to phase flashover, equipment damage, power failure, tower falling down, wire break and other serious accidents, so the early discovery and forecasting of conductor icing become important.Ice monitoring system is composed of two parts: the ice monitoring device and the upper manage system. This paper designs the overall program of icing and meteorological information remote monitoring system for transmission line, meanwhile studies the overall system architecture based on GPRS wireless networks and also with the data transmission and software modules in detail design. This paper focuses on the analysis and research of the methods of information processing of upper ice manage system, with the basis on the acquired weather information and icing image, it studies the models of icing and method of frequency analysis, and proposes a new calculation method of the thickness of transmission line icing based on image segmentation so as to estimate the severity of the icing, it uses improved level set algorithm with mixed functional model to extract the image edge and calculates the increased area, the final thickness and equivalent load of the conductor icing can be gained by comparing with the intrinsic size of the conductor area and conductor diameter, also the growth rate of the conductor icing will be known by calculating the difference of two ice thickness in a certain period of time, thus the situation and trend of transmission line icing can be evaluated correctly.Finally, this paper studies the detailed implementation of the monitoring system, including design of database and implementation of functional modules, moreover it completes analysis of icing frequency base on the sample of maximum ice thickness every year and gets the design value for transmission line, it calculates the ice thickness base on transmission line models using the icing images and the weather information obtained by the monitoring system, and finally it focus on the calculation of ice thickness based on image segmentation. Experimental results show that this algorithm can get a good segmentation of icing image with the advantages of high precision and low time cost. The result of icing thickness is closed to the actual value by comparing to the measured data which can correctly reflect the equivalent icing status of the conductor.更多还原