Abstract: With the vigorous implementation of the PV Poverty Alleviation Policy, a large number of PV load plants have access to the rural distribution network, in order to accurately and comprehensively assess the influence of the uncertain and correlative factors of distributed photovoltaic for rural distribution networks, probabilistic load flow calculations are applied for rural distribution networks that consider the relevance of photovoltaic output. Based on the characteristics of PV output correlation, a probability distribution model with correlated probabilistic input variables was constructed using in hybrid Copula theory, and a cumulant probabilistic load flow calculation method based on Copula theory was proposed to consider the correlation of input variables. Through the IEEE-33 bus system and an actual distribution network simulation in Gansu Province, the simulation results show that the model can display the probability characteristics of photovoltaic load generation well, and the hybrid Copula function is more accurate than the single Copula function in dealing with correlation problems. The results are compared with the Monte Carlo method to verify the rapidity and accuracy of the proposed method.