摘要
针对高压电气设备的高电压、封闭性和监测环境恶劣等特点,采用拉曼光纤传感器对其进行温度监测。根据故障特征量将故障进行分类处理,并利用多个并联的RBF神经网络进行高压电气设备故障的局部诊断,获得彼此独立的证据,再运用D-S证据理论融合算法对各个证据进行融合,最终实现对高压电气设备故障的准确诊断。通过实验证明:采用该诊断系统可有效提高诊断的可信度,减少诊断的不确定性。
High-voltage electrical equipment（HVEE） has the characteristics of high voltage and closure, and works in harsh environments. Raman fiber sensor was used for temperature monitoring. After faults were classified, and several shunt-wound RBF networks were used to local HVEE fault diagnosis, independent evidences were acquired. Then D-S evidence theory fusion algorithms were used to fuse evidences. Accurate diagnosis was achieved finally. The diagnostic tests prove that the system can improve the reliability of the diagnosis and decrease the uncertainty markedly.
出处
《河北科技大学学报》
CAS
2012年第6期501-505,共5页
Journal of Hebei University of Science and Technology
基金
河北省自然科学基金资助项目(F2009000728)
关键词
高压电气设备
故障诊断
拉曼散射
信息融合
D-S证据理论
RBF神经网络
high voltage electrical equipment
fault diagnosis
Raman scattering
information fusion
D-S evidence theory
RBFneural networks
