Using integrated generator diagnosis to perform condition based maintenance

To get the best generator diagnosis possible, multiple tests must be done but this is time consuming and not always possible. For a company such as Hydro-Québec (HQ), operating more than 350 generators, a diagnostic strategy had to be defined to be able to efficiently diagnose generators. A centralized database used in conjunction with a WEB based diagnostic application (MIDA) was developed at its research institute (IREQ) and implemented in HQ. The application provides a quick and global view of the condition of all units by using simplified health indexes. This information facilitates the decision making process when time comes to decide on maintenance actions or on life extension. The current paper summarizes the diagnostic approach used in MIDA with some results and data trending based on four case studies. With the MIDA application, the diagnostic information is easily accessible and continuously updated so that specialist can be more efficient to detect ongoing degradation before failure occurs and to carry out root cause analysis after a failure.