Loading...

Data Mining Approach for Fault Diagnosis of Power Systemby Wide Area Measurement

Zamani Dehkordi, Payam | 2012

594 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 43238 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Ranjbar, AliMohammad
  7. Abstract:
  8. Fast and accurate fault diagnosis has a significant role in the power system operation. For a power system to come back to the normal state, the precise fault diagnosis is necessary. The current practice in power system fault diagnosis needs operators in the control center to use data of PR and CB in order to infer the faulted section based on their experiences. Expert systems and artificial intelligence have been suggested as alternatives to the human expert experience and inference. Other model- and optimization-based methods have also been reported for fault diagnosis, provided that similar data is available from the SCADA system.Nevertheless, in large disturbances resulting in partial or complete blackout,malfunction of protection system has often been among the main suspects of the disaster. When PRs and/or CBs mal-operate or fail to operate when required in such situations, fault diagnosis becomes very complicated or even impossible by the existing methods. In the meanwhile, fast and accurate fault diagnosis by the control center is crucial to the restoration of the power system. Therefore, an alternative approach for fault diagnosis independent of the protection system is necessary. This thesis introduces a novel approach for power system fault diagnosis based on synchronized phasor measurements during the fault. The proposed method uses the available signals following the fault and thus can be trusted even in case of protection system failure. The proposed fault diagnosis algorithm shows a good accuracy for different fault impedance values. Moreover, fault diagnosis is robust against PMU measurement error and power system topology changes
  9. Keywords:
  10. Fault Detection ; Wide Area Measurement System ; Data Mining ; Phasor Measurement Unit

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC