Time-series models for reliability evaluation of power systems including wind energy

An essential step in the reliability evaluation of a power system containing Wind Energy Conversion Systems (WECS) using sequential Monte Carlo analysis is to simulate the hourly wind speed. This paper presents two different time-series models generated using different available wind data. Wind data from Environment Canada and SaskPower are used to illustrate these models. No assumptions or previously estimated factors are included in the models. In order to check the adequacy of the proposed models, the F-criterion and Q-test are used, and the statistical characteristics of the simulated wind speeds are compared with those obtained from the actual wind speeds. The proposed wind models satisfy the basic statistical tests and preserve the high-order auto-correlation, seasonal property and diurnal distributions of the actual wind speed. Copyright © 1996 Elsevier Science Ltd.