Based on the Copula function theory，this paper presents a multi-dimension correlated wind speed model and applies the model into a Monte Carlo simulation method for composite generation and transmission system reliability evaluation.Historical wind speed data is used to verify the method.It shows that wind speed correlation model offers better results for statistical properties of wind speeds including probability distributions，basic statistics and correlations than that of traditional matrix transform method.Case studies on the IEEE-RTS system with wind farms indicate that the traditional matrix transform method results in underestimation of composite system reliability index.