基于概率神经网络的直流滤波器电容器开路故障定位方法

针对高压直流输电工程中直流滤波器电容器开路故障定位方法效率低、威胁检修人员的安全等问题，提出了一种基于概率神经网络的直流滤波器电容器开路故障定位方法。该方法首先采用主成分分析法过滤冗余的信息，既降低了数据的维度，也得到了能有效反映直流滤波器电容器运行状态的特征向量；然后利用概率神经网络对特征向量进行故障分类，从而实现对电容器开路故障的定位。利用直流滤波器电容器运行状态样本数据对所提定位方法的有效性进行验证，同时考虑电容器有不同的开路故障电容元件数，验证所提定位方法的适应性。实验结果表明，文章所提方法对直流滤波器电容器的开路故障位置具有较好的定位效果。

Open-circuit fault location method of DC filter capacitor based on probabilistic neural network

Aiming at the low efficiency of DC filter capacitor open circuit fault location method in HVDC transmission project, which threatens the safety of maintenance personnel, a probabilistic neural network based DC filter capacitor open circuit fault location method is proposed. Firstly, the principal component analysis method is used to filter the re-dundant information, which not only reduces the dimension of the data, but also obtains the eigenvector that can effectively reflect the operating state of the DC filter capacitor; Then the probabilistic neural network is used to classify the feature vector, so as to realize the location of capacitor open circuit fault. The validity of the proposed location method is verified by using the sample data of DC filter capacitor operation state. At the same time, con-sidering the different number of capacitor elements with open circuit fault, the adaptability of the proposed location method is verified. The experimental results show that the proposed method has a good location effect on the open circuit fault location of DC filter capacitors.