This paper proposes Latin hypercube sampling (LHS) methods for reliability analysis of power systems including renewable energy sources, with an emphasis on the fluctuation of bus loads and intermittent behavior of renewable generations such as wind and solar power. The LHS methods that are applicable for systems with correlated random variables-system load and renewable generation-are proposed. Reliability indices such as loss of load expectation and loss of load probability are estimated. Results from Monte Carlo (MC) sequential sampling, MC nonsequential sampling, and that from the proposed LHS methods are compared. It is shown that the proposed methods are as accurate as the other sampling methods while requiring much less CPU time. Two case studies modified from the Electric Reliability Council of Texas (ERCOT) and IEEE Reliability Test System (IEEE RTS) are presented to demonstrate the performances of the proposed sampling methods.