电力变压器发生铁心松动故障后,虽然振动信号中基频分量(100 Hz)会发生变化,但影响振动基频信号的因素众多,很难据此准确诊断铁心松动故障。研究发现,变压器铁心松动后,50 Hz分量及其部分倍频分量、基频的倍频分量等新特征频率能够反映故障。根据特征频率之间的关系提出基于振动的变压器铁心故障诊断方法,不仅能够诊断出变压器铁心内部有无故障,还能判断其故障类型。为了验证提出的模型和方法,文中对实际变压器设置铁心松动故障,采集其监测点处的振动信号,经过消噪、折算处理,采用提出的诊断方法对实际变压器进行故障诊断,其诊断结果与实际故障一致,验证了提出的诊断方法准确、可行。根据所提出的诊断方法研制出基于振动的电力变压器故障诊断系统,并应用于实际变电站。
When a core loosing fault of power transformer occurs, the fundamental frequency component(100 Hz) in its vibration signal will change, but the fundamental frequency component cannot be used for accurate diagnosis of the fault because many factors influence the component. It is known that the new characteristic frequencies of the 50 Hz frequency component and some of its harmonic components,as well as the harmonic components of the fundamental frequency, can also reflect the fault of transformer core loosing.In this study, a fault diagnosis method of transformer core loosing was proposed based on the relationship among the characteristic frequencies. This method can diagnose whether a fault exists in transformer core, and determine the fault type.In order to verify the proposed method, a deformation fault was set to the actual transformer core. The vibration signals were sampled from the monitoring points, and de-noising and conversion processing was performed to diagnose the fault with the proposed method. The diagnosis result is the same with the actual fault, indicating that the proposed diagnostic method is accurate and feasible. Fault diagnosis system of power transformer based on vibration has been developed by using the proposed diagnosis method and applied to the actual substation.