【作者】 曹永刚；

【导师】 周玲；

【作者基本信息】 河海大学 ， 电力系统及其自动化， 2006， 硕士

【摘要】 电力变压器是电力系统中最重要的电气设备之一。随着我国电力事业的发展,电网容量的不断增大,用户对供电可靠性的要求不断提高,维修管理的重要性日益显现出来。维修费用占电力成本的比例也不断提高。在状态维修体制下如何选择合理的维修方案,以保证在不降低可靠性的前提下节省维修费用。 本文将概率神经网络应用到电力变压器的故障诊断中,解决了BP神经网络等神经网络训练时间长,样本变化时需重新训练等缺点,引入遗传算法确定概率神经网络的平滑系数,从而提高了诊断的正确率。本文首先介绍了灰色局势法、可拓优度法和TOPSIS法,利用模糊模式识别技术对灰色局势决策法进行了改进,并利用社会选择函数对三种决策方法得到的结果进行集结和重排序,形成电力变压器状态维修综合决策方法;其次从安全性、可靠性、经济性等角度出发建立了状态维修评价指标体系,并且采用层次分析法计算了各评价指标的权值;最后利用提出的电力变压器状态维修综合决策方法分析了两个不同的案例,仿真的结果证明这种方法是正确的。同时应用以上理论开发了电力变压器在线诊断与维修决策系统。更多还原

【Abstract】 The power transformer is one of the most important equipment of the power system. The power distribution reliability and safety of power system is directly affected by the transformer whether it can run or not. Maintenance control becomes more and more important when the capacity of power system becomes more and more large and the requirement of the power distribution reliability to the customers becomes more and more strict. The ratio of the maintenance costs in the power costs is increasing. Under the condition maintenance structure, the most important thing is how to choose rational maintenance program and exact maintenance schedule in order to save maintenance costs.A new method based on probabilistic neural networks (PNN) to transformer fault diagnosis is presented, the major features of the probabilistic neural networks stems from its modular architecture design and can be easily extended to adapt to a changing environment by incremental learning. The genetic algorithm is introduced to train the smoothing factor of PNN in order to increase the accuracy of diagnosis. Firstly, the improvement of grey situation decision, superiority evaluation, TOPSIS are introduced into this paper, and the grey situation decision method was been retrofitted by fuzzy pattern recognition. Using the social choice function, the power transformer condition maintenance synthetic decision-making method is formed by the consolidation and re-collation of the result of three decision-making method mentioned above. Secondly, the condition maintenance evaluation index structure is established including the requirement of safety, reliability, economy and others, the weights of the evaluation index are determined by analytic hierarchy process. Finally, two different instances are simulated by the power transformer condition maintenance synthetic decision-making method, and this method is proved to be correct. The power transformer online diagnosis and maintenance decision-making system is developed by the theory mentioned above.更多还原