变压器的运行状态直接影响到整个电网的运行安全，如何实现对变压器多种故障实时监测、保障电网的正常运行至关重要。本文提出基于多频超声波的变压器故障检测技术，即同时利用上百个频率不同的超声波对变压器油不间断扫描检测，并对接收到的超声波的各项参数采用多元统计分析技术和复数人工神经元网络数据分析技术进行处理，得到不同超声波的特征值，经过与变压器故障特征值比对可得到变压器的运行工况和故障名称，从而实现高效、准确、同时检测出多种故障。本文受中国南方电网科技项目资助（项目编号：GZGD20150301240091）。%This paper proposes an effective, acute and synchronous transformer fault detection method based on the principle of multi-frequency ultrasonic waves. This method utilizes hundreds of ultrasonic waves with different frequencies to continuously detect the transformer oil and handles the waves’ parameters with multivariate statistical analysis technology and plural artificial neural network data analysis technology. In this way, we can get the operational status and names of the faults by comparing the eigenvalues of the faults. The operational status of transformers may directly affect the safety of the entire power grid. Therefore, it is essential to monitor a variety of transformer faults and ensure the regular working of the power grid. This paper is supported by Science and Technology Projects of China Southern Power Grid (Project Number: GZGD20150301240091).
展开▼
