Assigning transition rates to unit models with incomplete data for power system reliability analysis

In power system reliability evaluation, one of the basic data needed is the transition rates between various states of components such as generators, transmission lines and circuit breakers. These data are used to build the transition rate matrix for the components which can then be used to find state probabilities and interstate frequencies. There are some situations when a transition rate matrix for a unit is needed but data is available only for the computation of state probabilities and data for transition rates is either not available, is incomplete or in error. An important case is that some commercial software, especially those using sequential Monte Carlo simulation need transition rates but the available data is either incomplete or lacking to determine transition rates but is sufficient to compute state probabilities. This is often a result of data keeping practices. This paper discusses this problem and proposes a solution for finding transition rate matrix for a component model when data is available to calculate the state probabilities but data for the estimation of transition rates is either incomplete or unavailable. The ideas can also be used to make corrections when the data is available but in error.