Font Size: a A A

Research On Service Restoration Of Actcive Distribution Network With Microgrids

Posted on:2018-12-11Degree:MasterType:Thesis
Country:ChinaCandidate:S Z ChenFull Text:PDF
GTID:2392330590977572Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
With large scale distribution generation(DG)connected to the distribution network,its high permeability has brought great influences,changing the supply mode and the power direction of distribution network,which lead to the problem of power supply reliability.Microgrid is the way to solve this problem with its controllability and reliability,and the research on the intermittent characteristics of DG in the microgrid is the basis of its power control.Therefore,the source stochastic power processing model based on superquantile is proposed to solve the problem of output prediction and random estmation.With the development of active distribution network(ADN),DG with high permeability begins to play an important role in service restoration,especially when electric vehicles(EV)are promoted widely,which causes a large number of electric vehicle station(EVS)accessed to the distribution network bringing widely distributed storage resources.Under this cricumstances,persistent problem of the recovery island in the time dimension finds a new way to solve through active control.Thus,the fault restoration model of active distribution network is constructed,and the recovery strategy of the short time islanding microgrid is proposed.Active distribution network is the advanced developing stage of intelligent distribution network.In the future,it is normal for the distribution network to integrate microgrids with all kinds of DGs.The effective use of multi-microgrid for recovery after the fault,as well as the interconnection of multi-microgrids and other issues,has become the focus of research.For this reason,the thesis proposes a multi-agent based fault recovery scheme for intelligent distribution network with multi microgrids.The specified achievements of the thesis are as follows:Firstly,the formation,status of operation and evaluation index of the island with DG or microgrids in ADN are illustrated.Secondly,the stochastic power processing model based on superquantile estimation is proposed.Then the available vehicle quantity probability model of EVS after the failure is built.Also,the load characteristics,charging and discharging strategy of EV and the quantity model solving process are analyzed.And the output probability model of wind turbine and solar photovoltaic is concluded.The proposed model is applied to solve them.Then,the islanding restoration strategy of ADN based on multi-source Steiner tree is proposed.To start with,the ADN restoration double-constrainted model is constructed,including the upper island partitioning layer for maximum available capacity planning and the lower path search layer based on dynamic fuzzy set.And the island partitioning algorithm is constructed based on core tree and multi-source Steiner tree algorithm.With its encoding and criterion improved,the IEEE33 network is tesifed by improved binary particle swarm optimizaiton.Moreover,a restoration framework and its process based on multi-agent system for intelligent distribution network with multi microgirds is proposed.Lastly,the key self-healing technology of intelligent distribution network with multi microgrids is analysed,including the restoration characteristics of traditional centralized control mode,analysis of the key technical problems for multi microgrids interconnection and relay protection problems brought by it.
Keywords/Search Tags:active distribution network (ADN), microgrid, superquantile, multi-source Steiner tree, multi-agent system
PDF Full Text Request
Related items