多传感器信息融合技术在变压器故障诊断中的应用

Application of Multi-Sensor Information Fusion Technology to Transformer Fault Diagnosis

本文中作者通过采集变压器运行时的状态参数,运用并行的两个BP神经网络对变压器进行局部诊断,再用D-S证据理论对局部诊断的结果进行全局融合,实现了对变压器故障的准确诊断。

The parameters of transformer in operation are integrated by anthors in this papers. Using two parallel BP neural networks to transformer in partial diagnosis, garnish with D-S evidence theory to partial diagnostic results,the authors realize accurate diagnosis of transformer foult.