论文部分内容阅读
随着电力系统的规模越来越大,其自动化要求程度也越来越高,现在电力系统中许多变电站为实现无人值班或少人值班的目的,大都使用了微机监控系统来对有载调压变压器OLTC(On—Load Tap—Changer Transformer)分接头的位置和电容器组的投切进行监视与控制。但是在实际的运行过程中,却出现控制器动作频繁、波动大的问题,最后不得不将自动改为手动,失去了自动化作用。本文针对这种情况,在变电站采用电压——无功功率双参数调节控制策略的基础之上,引入了人工神经网络ANN(Artificial Neural Network)中的海明网络(Hamming)来对此控制策略进行模式差别,以提高控制器动作的准确性,从而达到减少动作次数,实现自动控制的目的。 传统的变电站控制策略是单一参数电压的控制,仅是根据电压的高低来进行判断,根本不涉及无功功率补偿的问题,而采用电压——无功功率双参数控制,则是在根据电压控制的基础上,引入无功功率的调节,将有载调压变压器分接头的位置与电容器的投切结合起来,利用电压和无功功率之间的强相关性,来达到改善电压质量和降低线路损耗的目的。根据电压——无功功率双参数综合控制原则,将控制区域分为九区,分类如图1所示。
With the increasing scale of the power system, the degree of automation requirements is also getting higher and higher. Nowadays, many substations in the power system have mostly used the computer monitoring system to carry unloaded or less people on duty. Pressure Transformer OLTC (On-Load Tap-Changer Transformer) tap position and capacitor bank switching to monitor and control. However, in the actual operation process, but there are frequent controller movements, large fluctuations, the last had to be automatically changed to manual, lost the role of automation. In view of this situation, based on the voltage-reactive power two-parameter regulation control strategy in substation, Hamming, an artificial neural network (ANN), is introduced to make this control strategy Model differences in order to improve the accuracy of the controller action, so as to reduce the number of actions to achieve the purpose of automatic control. Traditional substation control strategy is a single parameter voltage control, only based on the voltage level to judge, does not involve the problem of reactive power compensation, and the use of voltage - reactive power two-parameter control, is in accordance with the voltage control , The introduction of reactive power regulation, the position of the on-load tap-changer tap with the capacitor switching, the use of strong correlation between voltage and reactive power, to achieve improved voltage quality and reduce the line The purpose of loss. According to the voltage - reactive power two-parameter integrated control principle, the control area is divided into nine areas, classified as shown in Figure 1.