Time filter

Source Type

Guo C.-X.,Zhejiang University | Zhu C.-Z.,Hangzhou Electric Power | Zhang L.,Northwest China Grid Company | Peng M.-W.,Zhejiang University | Liu Y.,Zhejiang University
Zhongguo Dianji Gongcheng Xuebao/Proceedings of the Chinese Society of Electrical Engineering | Year: 2010

A novel support vector machine (SVM), i.e. multiclass multiple-kernel learning support vector machine (MMKL-SVM), for the fault diagnosis of power transformers is proposed in this paper. Unlike traditional SVM that may fail under some circumstances, the fault diagnosis method based on MMKL-SVM has some good theoretical properties, e.g. it only deals with a simple objective function, and the classification results can be obtained by direct calculation on the basis of a simple decision function; it can conduct calculation with an optimal kernel function composed of linear combinations of basic kernels, further boosting the overall performance; the solutions for it can be efficiently gained by iteratively solving two convex optimization functions with a low computation cost and high speed. Diagnosis test results show that the MMKL-SVM method has high classification accuracy, which proves its effectiveness and usefulness. © 2010 Chin. Soc. for Elec. Eng. Source

Suonan J.,Xian Jiaotong University | Wang C.,Xian Jiaotong University | Kang X.,Xian Jiaotong University | Ma C.,Xian Jiaotong University | Liu X.,Northwest China Grid Company
Zhongguo Dianji Gongcheng Xuebao/Proceedings of the Chinese Society of Electrical Engineering | Year: 2013

To eliminate the capacitive current's influence on line protection of long-distance and high-voltage transmission lines, this paper proposed a novel transmission line protection scheme based on model identification method and on the point that matrix pencil algorithm can extract the characteristic frequencies of the transmission line reliably. By studying the fault network topology of distributed parameter lines, the frequency-domain models of external and internal fault states were built. The fault model for external fault states is a distributed capacitive model, while the fault model for internal fault state is an equivalent system impedance model. The two kinds of unbalanced currents were defined as two model errors to identify different fault models. When an external fault occurs, the fault data corresponds with the distributed capacitive model and its unbalanced current is smaller than that of the equivalent system impedance model. When an internal fault occurs, fault data corresponds with the equivalent system impedance model and its unbalanced current is smaller than that of the distributed capacitive model. Therefore, external and internal faults are distinguished. Theoretical analysis and simulation results show that the new principle can remove internal fault quickly and reliably, free from the influence of capacitive current and transition resistances. © 2013 Chin. Soc. for Elec. Eng. Source

Su P.,Northwest China Grid Company | Zhang K.-S.,Xian University of Technology
Dianli Xitong Baohu yu Kongzhi/Power System Protection and Control | Year: 2010

The LVRT control strategy and protection schemes of DFIG are discussed in condition of large external voltage dip. On the basis of analyzing active IGBT Crowbar circuit topology as well as the effect of the Crowbar circuit to the fault during grid voltage dip, the DFIG mathematical model with considering the changed grid voltage is adopted and the control model of LVRT is established. The switching strategy of Crowbar is researched in detail during simulation. The simulation results verify that Crowbar circuit and the control strategy are active and prove that Crowbar control strategy could limit the over current in the rotor and the over voltage of the DC bus as well as the transient oscillation of the electromagnetic torque efficiently. Moreover, a reactive current is injected into the grid to assist the recovery of the grid voltage, which is in favor of achieving the LVRT of DFIG. Source

Northwest China Grid Company | Date: 2015-01-09


Northwest China Grid Company | Date: 2015-01-09


Discover hidden collaborations