Wind energy conversion is the fastest-growing source of renewable electric power generation in the world and it is expected to remain so for some time. Doubly Fed Induction Generators (DFIGs) hold more than 50% market share of wind power generators due to its unique benefits of low cost, control capability and scalability. Recent trend of DFIGs installed in wind power are within power range of 200KW-2MW and have stator windings rated at 690V, which are highly susceptible to inter-turn faults. This fault type account for more than 40% of the total failure causes of the generator. A reliable fault diagnostic system to detect this fault at incipient stage is highly in demand to prevent catastrophic failures and reduce preventive maintenance costs, since regular inspection is expensive and rewinding of generator takes time in period of days and costs nearly 20% of the original generator cost. Conventional techniques developed for detecting stator inter-turn faults in induction motor are not directly applicable for this machine partly due to difference in construction and their operating characteristics. Research for fault diagnosis in last 5 years has lead to variety of techniques, but none have proved to be feasible either due to need for dedicated sensors and high computing needs or they fail to account for operating conditions which lead to ambiguous fault triggering. This absence of any proven technique to detect stator inter-turn faults in DFIG and the dire necessity for a fault diagnostic system to promote their reliability in wind power generation has motivated the research of this thesis.
A novel technique to detect stator inter-turn faults in a DFIG is proposed by analyzing rotor current and search coil voltage in this thesis. Our recent observations suggested that harmonics induced in the rotor circuit are very promising in detecting stator inter-turn faults in DFIGs. Hence. in this study, an in-depth investigation was conducted to determine the origin of various harmonic components in the rotor currents and their feasibility to detect inter-turn stator faults unambiguously. The theory is verified by modeling the DFIG and simulating various operating conditions of healthy state. severity of faults and unbalance loading conditions that cause ambiguity. Further to verify the concept, an experimental test-bed was created and tests were carried out on DFIG under varying speed operations, fault severity and loading conditions. The experimental results achieved from analyzing the frequency spectrum of rotor current and search coil voltage have been compared and discussed.
To verify feasibility of the proposed technique for detecting stator inter-turn faults and obtain results on speed sensitivity of fault detection, a prototype of DSP based Fault Diagnostic device was developed. Further discussion is made on challenges of realizing the proposed fault diagnostic technique in the industry and scope for further improvement.
| Identifer | oai:union.ndltd.org:uvic.ca/oai:dspace.library.uvic.ca:1828/2589 |
| Date | 13 April 2010 |
| Creators | Shah, Dhaval |
| Contributors | Nandi, Subhasis |
| Source Sets | University of Victoria |
| Language | English, English |
| Detected Language | English |
| Type | Thesis |
| Rights | Available to the World Wide Web |
Page generated in 0.0021 seconds