Geometric-Based Fault Diagnosis in a Grid-Connected Inverter

Mashreghi, Ali | 2021

38 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 53838 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Namvar, Mehrzad
  7. Abstract:
  8. Inverters are the main utility in transmitting power from distributed generation (DG) systems toward grid and their safe and immune performance has become a major concern among the system designers. In addition to poor power transmittance, faults in inverter can harm other equipment such as DG and local load. Consequently different protections has been considered to increase the reliability of these systems. Some of these protections like fuses and varistors are implemented in the hardware while others are considered in the processor. In the later class of protections, fault detection and isolation (FDI) system has been the subject of intense investigation in inverters over the past two decades. Nonetheless available methods have shortcomings in: exploiting additional sensors, sensitivity to load, threshold tuning, additional assumptions on system signals and ... so that they can lead to false alarms or additional costs. Moreover, investigating both sensors and power switches' faults in a single scheme is challenging and has been rarely discussed. In this thesis with the aid of geometric tool a fault detection and isolation scheme is designed for nonsimultaneous faults in inverter power switches together with faults in grid voltages and source current sensors. This scheme addresses previous mentioned challenges and its performance is proven in simulations and practical tests. In this regard a comprehensive algorithm is introduced for FDI system design in input affine systems with nonsimultaneous faults
  9. Keywords:
  10. Geometric Approach ; Grid Connected Inverter ; Sensor Error ; Power Switch Fault ; Fault Detection and Isolation (FDI) ; Distributed Generation

 Digital Object List