The role of availability parameters for the choice of stand-alone power plants

The paper tackles the problem of the optimal sizing of a power system at the planning stage-that consists in the best choice among many alternatives-as a function of reliability and maintenance, both encompassed in the availability function of the system. Moreover, statistical methods to deal with uncertainty of such parameters are also discussed and illustrated with reference to typical values of wind turbines. In particular, the case studies refer to the sizing of generation capability of a power system, that consists in the choice of the optimal number and size of the generators (assumed as having all the same size and availability) to be installed for a given total installed power. The results show that the solution that features the highest availability might not be also the best in terms of expected performances, evaluated e.g. as expected energy not supplied. This shows that the introduction of proper ldquorisk indexesrdquo is essential for characterizing the overall performances of the system. Moreover, under adequate assumptions, the treated indexes can be expressed analytically in terms of parameters relevant to reliability and maintenance, i.e. mean-time-to-failure (MTTF) and mean-time-to-repair (MTTR); this enables a proper sensitivity analysis of the same indexes with respect to reliability and maintenance parameters to be performed. Finally, some numerical applications are illustrated regarding a new statistic methodology for handling the unavoidable randomness of the above availability parameters, in order to perform a more realistic and rational choice among different alternatives.