Reliability Models to Estimate Mean Time to Failure of Switched Reluctance Machines

In this paper, reliability models for Switched Reluctance Machine (SRM) are presented to estimate Mean Time to Failure (MTTF) when operated during normal and fault conditions. Markov state transition diagrams and reliability models are used to estimate the reliability of the machine following different fault occurrence levels. The variation of phase currents, output torque, and magnetic force acting on stator and rotor poles due to fault in the machine are recorded for failure rate estimation from electromagnetic simulations. The failure rate of each part in SRM is calculated using empirical-based models and is combined to calculate the machine's reliability and MTTF. A 32 kW, 18 slot/12 pole SRM is considered to perform state assessment and calculate the failure rate of parts in SRM. For validation of the presented results, reliability prediction from phase currents and torque obtained from the experiment is compared with simulations.