A generation system reliability model with Monte Carlo simulation

There is a growing need within electric utility systems operations to refine approaches for assessing reliability parameters such as loss of load probability (LOLP) and expected unserved energy (EUE). These parameters are affected by the system operation policy requirements such as spinning reserve constraints, the minimum unit up/down and starting times. Account of those requirements renders the methods like Baleriaux-Booth convolution impossible. In our opinion Monte-Carlo simulation of the generating units availability is preferable for the calculation of the reliability parameters when considering the system operation policy. An important fact is that the availability status of a generating unit is dependent on its current availability state. Therefore the use of Markov transition matrices in approximation of electric utility system operation is more adequate than the use of the partial forced outage probabilities. Obviously, both unit commitment plans and forced outages of generating units have a direct effect on the value of available system capacity.