CEEMDAN与GCN结合的配电变压器故障诊断

针对配电变压器故障特征提取困难、故障识别难度大的问题,提出一种将振动信号、自适应噪声完备集合经验模态分解(CEEMDAN)与图卷积神经网络(GCN)三者有机结合的故障诊断方法。 首先,采用 CEEMDAN 对来自加速度传感器的振动信号进行处理,获得一组固有模态分量(intrinsic modal function);其次求取边际谱信息作为特征向量;然后,对特征向量矩阵构造无向加权完全图,并使用改进灰狼优化算法对高斯核带宽进行寻优;最后,搭建一个具备多通道和多连通的改进 GCN 模型进行特征二次挖掘与故障分类。 与此同时,还在模型中加入一种名叫“峰值因子”指标实现对未知类型故障的辨识。 在实例分析中,分别对油浸式和干式变压器进行故障模拟,提取不同状态的样本进行测试。 实验结果表明,所提方法对油浸式和干式变压器的故障识别准确率分别达到 97. 73%和 95. 6%,优于其他两种对比方法。 在面对未知类型故障以及运行工况发生变化时,也具备较高是识别能力。

Fault diagnosis of distribution transformer based on CEEMDAN and GCN

Aiming at the difficulty of fault feature extraction and fault identification of distribution transformers, a fault diagnosis methodcombining vibration signals, complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and graph convolutionneural networks (GCN) was proposed. Firstly, the vibration signal from the acceleration sensor is processed by CEEMDAN to obtain aset of intrinsic modal functions. Secondly, its marginal spectrum information is taken as the feature vector. Then, an undirected weightedcomplete graph is constructed for the eigenvector matrix, and an improved gray wolf optimization algorithm is used to optimize theGaussian kernel bandwidth. Finally, an improved GCN model with multi-channel and multi-connectivity is built for feature secondarymining and fault classification. At the same time, an index called peak factor is added to the model to realize the identification ofunknown faults. In the case analysis, the fault simulation of oil-immersed transformer and dry transformer is carried out respectively, andsamples of different states are extracted for testing. The experimental results show that the accuracy of the proposed method for oilimmersed transformer and dry transformer fault identification is 97. 73% and 95. 6%, respectively, which is better than the other twocomparison methods. In the face of unknown types of faults and operating conditions change, it also has a high ability to identify.