DIKUL - logo
E-viri
Celotno besedilo
Recenzirano
  • Online Condition Monitoring...
    Fu, Qiang; Zhu, Jiaojiao; Mao, Zhi-Hong; Zhang, Guixin; Chen, Tefang

    IEEE transactions on industrial electronics (1982), 04/2018, Letnik: 65, Številka: 4
    Journal Article

    According to the special working condition in the commutation region of the traction grid, a novel technique of online core-loss measurement for the electric locomotive onboard traction transformer is proposed to noninvasively monitor the status of the magnetic core in real time. Meanwhile, a new modeling approach is also put forward for the equivalent nonlinear magnetizing inductance <inline-formula><tex-math notation="LaTeX">L_{M}</tex-math> </inline-formula> of the magnetic core. First, a no-load test under 31 kV is implemented on a healthy traction transformer, and the measured transient terminal voltages and currents are then filtered and downsampled by the comprehensive test and fault diagnosis system of electric locomotive traction transformer. Next, the magnetizing inductance <inline-formula><tex-math notation="LaTeX">L_{M}</tex-math></inline-formula> is modeled as the ratio of "the derivative of main flux linkage λ p with respect to time" over "the derivative of magnetizing current <inline-formula><tex-math notation="LaTeX">i_{m}</tex-math> </inline-formula> with respect to time." After that the proposed <inline-formula><tex-math notation="LaTeX"> L_{M}</tex-math></inline-formula> model is verified by comparing with the experimental results and two other models. Finally, the proposed technique for online estimation of core loss is applied on a normal onboard traction transformer and a downscaled testing transformer with different severity of interlaminar fault. The experimental results approximately match the theoretical calculation, and the online estimated values of core loss show sharply rising trend under incipient fault.