Reliability models of wind farms considering wind speed correlation and WTG outage

•Both wind speed correlation among different wind farms and wind turbine generator's forced outage rate (WTG FOR) are considered.•The correlation of wind speed and WTG FOR is considered.•A new approach based on binomial distribution and random sampling is proposed for the consideration of WTG FOR.•A linear apportioning method is proposed to create multistate models of wind farms based on the hourly output power model of wind farms.•The effect of dependency between wind speed and WTG FOR is minimal when WTG outages are independent, but substantially larger when WTG outages are dependent.

Wind speed correlation and wind turbine generator (WTG) outage are two factors affecting the reliability model of wind farms, but they are not addressed simultaneously in the existing literature. Meanwhile, WTG outage is reported to be dependent with wind speed to some extent. Therefore, the extended reliability models of wind farms incorporating both of these two factors and the dependency between WTG outage and wind speed are proposed in this paper. To consider the uncertainties and dependencies of wind speed and WTG failure, Copula method is applied to simulate correlated random variables representing for wind speed and the number of failed WTG units. Moreover, the linear apportioning technique is used to create multistate reliability models of wind farms from hourly wind power models. A number of sensitivity analyses on the modified IEEE RTS with wind power are conducted to validate the proposed reliability models for generation adequacy assessment. Case studies show that the generation adequacy indices increase with the correlation of wind speed and WTG forced outage rate (FOR). It is meaningful to point out that the effect of dependency between wind speed and WTG FOR on generation adequacy is minimal when WTG outages are independent, but it will be substantially larger when WTG outages are highly dependent. The proposed multistate reliability models of wind farms provide foundation for the reliability assessment of power systems with wind power integrated.