Study On Reliability Assessment Method Of Distribution System Containing Distributed Generation

Distribute generation represented by wind power and photovoltaic power is increasingly used in the distribution network with the development of economy and the progress of science and technology. Complementary power supply system containing the output of wind and photovoltaic power changes the original distribution network topology structure directly, power supply mode of the distribution network is changed from a single power supply to multiple power supply, the original distribution network reliability evaluation method has not adapt to the distribution network containing the wind and photovoltaic power. Because the distribution network occupies an important role in the large and complex power system, so studying the reliability evaluation of distribution network containing the wind and photovoltaic power has an important practical significance.Studying the probability distribution model of wind speed and illumination data is the preparation of evaluating the reliability of the distribution network. In order to build a probability distribution model closing to the physical distribution of wind speed and illumination data, the paper uses the traditional probability distribution model and the nonparametric kernel density estimation method to fitting the wind speed and illumination data. Through the comparison, the nonparametric kernel density estimation method is considered to be the more method of closing to the physical distribution of wind speed and illumination data.Most of reliability evaluation methods of the distributed power distribution network think distributed power output as a fixed value, the paper establishes a binary joint copula function which reflects the correlation of wind speed and illumination data. In combination with the output of wind power and photovoltaic power system value per hour, the paper gets the output model of multiple states of distribution network containing the wind and photovoltaic power.In the problem of island partition, this paper constructs the target model of island, and adopts the depth-first search algorithm to divide the island area, then search each load by the importance of each load. This model can restore most of the loads when the power distribution network is fault and meets the demand of important load of power supply.Finally, the paper puts forward the reliability evaluation method based on the sequential monte carlo simulation method of distribution network containing the wind and photovoltaic power. The paper tests the effectiveness of the proposed method through IEEE- RBTS BUS6 system.