Probabilistic analysis of distribution network considering nonlinear correlation random variables

This paper analyses the stochastic dependence between random variables in the distribution network containing a variety of distributed generations, and obtains the nonlinear correlation properties between different random variables of wind power generation, photovoltaic generation and solar-wind hybrid power system. A joint probability distribution model based on the Copula theory is presented to transform the joint probability density into marginal probability density functions and Copula function, and two-stage method is used to analyse and choose the Copula function. The process combining the Copulas theory and Monte Carlo simulation based on Latin hypercube sampling is proposed to analyse the influence of correlated random variables on the node voltage and three-phase unbalance degree in distribution network. The case study demonstrates that the established model can properly describe the joint probability characteristics of different distributed generations, and can comprehensively reflect the impact of randomness of the distributed generations on distribution network.