Abstract: Wind power and photovoltaic generation in large scale is connected to the distribution network in the form of distributed power supply,which increases the complexity in the reliability assessment of the distribution network.A multicapacity state model of the generator based on the Markov process is developed to simulate the output fluctuation in face of the randomness of the distributed power output.In order to avoid the over-estimation problem of the interval minimum method,an affine minimal path method is proposed to effectively reduce the range of calculation results.This algorithm considers not only the original parameter uncertainties for components and the load points in the distribution network,but also the correlation between the uncertain variables.With the IEEE-RBTS BUS 6power distribution system as an example,the rationality of the proposed model and algorithm is verified,and the influence of transient fault on the reliability evaluation of the system is analyzed.The result shows that the affine minimal path method has higher accuracy than the interval-minimum method in calculating the reliability index.The greater the uncertainty of the system is,the more obvious will the effect of interval reduction appear,which provides a reliable basis for reliability evaluation and system planning and design in actual engineering.