Font Size: a A A

Research On Islanding Partition And Reliability Planning For Active Distribution Network

Posted on:2018-06-25Degree:MasterType:Thesis
Country:ChinaCandidate:J H GuoFull Text:PDF
GTID:2322330533466759Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
As the current international and domestic energy system and economic system is reforming profoundly,in order to cope with the increasing power load and traditional fossil energy shortage,the new energy generation technology has become a hot research topic.With the development of renewable energy technology,the distributed generation(DG)are more and more connected to the distribution network with the advantages of flexible generation technique and environmental friendly.Implementing the active management and active control technology,they constitute the active distribution network(ADN).DG can supply power nearly to form the island.The reasonable operation mode of island can bring their superiority into full play so that improve the electric reliability,which is attracted by extensive attention of electric industry.The main work and achievements are as follows.1)Reactive power balance is an important factor restricting the feasibility and safety of islanding.As the main reactive power source of the active distribution network,the power characteristics of the distributed generation(DG)are often neglected so that it is difficult to satisfy the power balance of islands.In order to conciliate accuracy and computational performance,a mixed integer programming model for islanding partition of the active distribution network is established.The P-Q capability limits of DGs are linearized based on the analytic geometry model,which can consider the power characteristics of the doubly-fed wind generator,the synchronous generator,and the voltage source inverter-based generator.The product terms between the voltage and the power in the Kirchhoff's current equations are linearized using the bi-linear model.The separation model is used for solving the variations of the islanding network topology and the load shedding problems.The square terms are linearized by adopting the special ordered sets-2.The study cases are performed to verify the proposed model.2)Aiming at the multi-step and complicate problem of distribution network fault recovery and the independent operating of island,an ADN recovery method based on resilience strategy is put forward to improve the system robustness.The proposed method can be divided into two stages.In the first stage,an MIQCP model is designed-the blackstart DG adopt the droop control connected to the network,supporting voltage and frequency locally and adjusting the power output.The downstream system is split into several self-adequate islands.Compared with the traditional islanding partition model,the proposed model considers the dynamic varied frequency;in the second stage,the islands in a stable situation interconnect orderly through the heuristic search self-healing strategy and form the interconnected system.The study cases demonstrate effectiveness of the proposed strategy.3)Aiming at the limitation of time consuming and experience dependent in the two step of the distribution network planning search and verification,the research on multi periods ADN reliability planning is conducted on the basis of islanding partition model.The hierarchical model regards the ADN framework and DG location as integer decision variables,establishes the master problem with the object of minimum comprehensive cost,and obtains the network frame.The slave problem obtains the minimum charge of electricity shorage by conducting the N-1 criterion checking.The power flow constraints,branch capacity constraints and voltage amplitude constraints are considered in both problems.The optimal objective network frame is obtained by iterating the master problem and slave problem based on the generalized Benders decomposition algorithm.The proposed model is testified by study cases.
Keywords/Search Tags:active distribution network, islanding partition, distributed generation, reliable programming, Benders decomposition algorithm
PDF Full Text Request
Related items