| In the context of the country’s further energy transition and the development of non-fossil energy,more and more distributed power sources have begun to be connected to the grid.Although it has effectively reduced the growth of carbon emissions,the random uncertainty of its output has increased the complexity of the distribution network structure,while making the distribution of failure probability in different regions more variable and power outage reconstruction strategies more diverse.It also further poses new challenges to the increasing expectation of power supply reliability of smart distribution networks for power users.Therefore,in response to the above problems,the thesis conducts research from the following three aspects:Firstly,taking the actual data of the distribution network of each city in H province as an example,by comparing the number of failures throughout the year,spring,summer,autumn and winter and their cumulative probability on the single logarithmic coordinate,it is proved that it has self-similarity;Through the random fuzzy processing of the probability distribution parameters of the number of failures,it is more realistic to describe the probability distribution of the distribution network failures;according to the information of the work order,a fault level,the duration of the power outage,the number of sensitive users and the affected users are constructed.A large number of power outage risk assessment index systems for distribution networks use the entropy method to assess the risks of the distribution network,thereby providing support for the formulation of scientific,reasonable and effective fault repair strategies.Secondly,based on the analysis of the random and fuzzy distribution of the probability of daily failures in the region and the risk of power outage based on the second chapter,the actual situation of the "closed-loop design and open-loop operation" of the distribution network is further considered,and a safety consideration is proposed.The criterion’s random fuzzy wind and solar capacity allocation model takes into account the uncertainty of wind power output limited by wind speed and photovoltaic output affected by light intensity.Based on the evaluation of the system safety indicators such as the power of each branch and the node voltage and the fuzzy chance measurement,respectively Taking the maximum expected value of annual profit and the minimum expected value of effective reduction of wind and solar output as the objective function for joint solution,a random fuzzy solution for the constant capacity of wind and solar in this scenario is obtained.Thirdly,on the basis of Chapters 2 and 3,further consider the reconstruction problem of wind and wind access distribution network after power outage,and propose a random fuzzy reconstruction model considering wind and wind;in order to make the distribution network structure high in reliability and small in loss,The optimization objective function of the system network loss and the minimum risk of power failure are used to solve the optimization objective function of the grid reconstruction;at the same time,the optimization is solved by the NSGA-Ⅱ algorithm,and the particles are initialized by the rules based on the ring network coding during the solution process.For population,the constraint conditions are added to the coding,which further improves the operation efficiency.In the context of the large-scale grid connection of distributed power sources,this thesis provides for the reliability analysis of a high proportion of wind-solar power systems from failure risk assessment to distributed wind and solar power configuration capacity configuration to system reconfiguration strategies after failure.reference. |