基于概率神经网络的小电流接地系统模式识别故障选线方法及应用-控制理论与控制工程专业论文.docxVIP

摘要 摘要 万方数据 万方数据 摘 要 小电流接地系统是指中低压配电网络中，采用中性点经消弧线圈接地、经 大电阻接地或者中性点不接地等运行方式的系统。通常情况下，小电流系统发 生故障后，故障电流十分微弱，不容易被检测到，这对故障选线问题提出了巨 大挑战。目前小电流接地故障选线方法主要分为三类：一是利用故障稳态特征 分量进行选线，二是利用故障暂态特征分量进行选线，三是注入特殊信号进行 选线。这些方法在实际应用中均取得了一些成效，但从选线的准确率和稳定性 上来说还远远不够理想。 概率神经网络是一种可用于模式分类的神经网络，在机械、材料、环境工 程乃至经济领域有着较好应用，而在配电网故障选线上很少有人尝试。经过反 复研究和探索，寻找到配电线路零序电流小波能量、有功分量和五次谐波分量 三种故障特征量作为选线的依据。对故障模式进行了合理定义，重点突破了概 率神经网络中多种故障特征量有效融合的问题，提出一种基于概率神经网络的 模式识别选线方法。 通过对小电流接地系统模型进行大量 Matlab 仿真试验，研究了故障位置、 接地电阻、故障合闸角、中性点接地方式、配电线路结构和噪声干扰对故障特 征量的影响，收集了大量故障数据样本。同时对电弧高阻接地、混合线缆配电 网、噪声干扰下的接地故障以及中性点不同接地方式系统的概率神经网络模式 识别选线进行了广泛试验，验证了该方法具有较好的通用性和抗干扰能力。将 该方法与概率神经网络单一特征量选线和 BP 神经网络故障选线方法进行了比 较，证明了该方法具有准确率高，操作简单快速，故障知识丰富，易于拓展知 识库等特点。最后提出了一套本方法的故障选线装置设计方案。 关键词：小电流接地系统；故障选线；概率神经网络；模式识别；数据融合 I Abstract ABSTRACT The non-effectively grounded system mainly adapts the way of isolated neutral, neutral grounding by the arc extinction and neutral grounding with resistance in our country of low voltage distribution network. Once there is a failure with this kind of system, the faulted current is too small to be detected, for the current can not find its loop. This way can effectively protect the distribution network working stability and sustainability, but there are so many problems in the faulted line selection. At present the small current grounding faulted line selection method is mainly divided into three types: one is to make use of the faulted steady state features, the second is to use the faulted transient features, the third way is to inject special signal into the system to do the faulted line selection. As the fields of electronics, communication, computer and intelligent control have been making such rapided development of technology, people began to explore the use of more advanced methods and computer equipments in order to get a better effect. Artificial intelligence, neural network and fuzzy recognition are applied to the emerging technologies in the faulted line selection. Probabilistic neural network is a kind of neural