Reliability assessment of power system reveals the level of risk and conducts risk identification via quantitative diagnosis of the impact of uncertainty factors on power system operation. Therefore

the modeling accuracy of uncertainty factors has become the foundation and application premise of reliability assessment of power system. A semi-parametric Gaussian mixture model (GMM) for multiple random variables

which combines the advantages of parametric and non-parametric model

is proposed with both accuracy and computational efficiency taken into account. Based on the GMM

arbitrarily distributed continuous random variables can be modeled flexibly and efficiently and the correlation between variables is conveniently considered. A two-stage complex sampling method is employed to generate random samples of GMM in Monte Carlo simulation of reliability assessment. With the probabilistic modeling of electric load and wind speed as an example

IEEE-RTS79 is used to demonstrate the accuracy and feasibility of the proposed model.