| With the economic and social development,the users' higher and higher demands on the reliability of power supply are required.Distribution automation system is an important means to improve the reliability of power supply,and has gradually become a hot issue at home and abroad.But most scholars did not make a comprehensive evaluation of economic costs and reliability,and didn't considered the actual situation that the appropriation and construction are accomplished stage by stage.Therefore,it's meaningful to optimize the investment plan,quantity configuration,sites and types selection of the distribution automation terminal units,and to achieve the greatest technical and economic benefits.For this purpose,the following work is done in this thesis:Firstly,an optimization method of load transfer is proposed.The optimization variables are the opening and closing states of the switches,and the optimization goal is the sum of the absolute difference between the injected power of the node before and after the power outage.Tree topology constraints,power flow constraints and other threshold constraints are taken into consideration.Thus,a nonlinear mixed integer programming model is established.The model is used to calculate the minimum power outage area,and to obtain the nodes that need to be isolated,transferred or repaired because of the fault and used to calculate the reliability index.A method to optimize the configuration of the distribution automation terminal units is proposed.This method considered both power supply reliability and economic cost.Different kinds of terminal units are taken into consideration.Annual outage time of per user caused by fault is chosen as the index to evaluate the power supply reliability.Models with different optimization goals are established,and these goals are respectively to optimize reliability index,total investment and output-to-input ratio.These models are used to evaluate the technical and economic benefits of the configuration scheme.The dynamic programming model to optimize the investment and configuration of the terminal units is established.The optimization goals are the annual outage time of per user caused by fault and the sum of power loss.The investment plan and configuration calculated by the models can not only increase social benefits from the point of users,but also improve economic benefits for the grid.And it's convenient to be implemented in engineering project if the appropriation and construction are accomplished stage by stage.Genetic algorithm is used to solve the optimization models.The individual fitness,coding method,parameter setting and genetic operators for the models of this thesis areexplained.And how to use Jenetics(an open-source JAVA library for genetic algorithm modeling)to solve the optimization models is introduced in detail.This algorithm is easy to be understood and implemented,and has a good computing performance.JAVA language is used for programming.The distribution data is read from the XML file provided by power utilities.Network parameters and topology information are stored with JGraphT(another open-source JAVA library for graph theory).Then the optimization models are established based on simplifying the network,and the results of the models are obtained by Jenetics.The code can be packaged into executable jar package,which can run in different environment,and can be imported into other related projects to facilitate the integration of the projects.Finally,the actual data of a city's distribution network is used as examples.And The examples in this thesis are compared with the methods of three-section configuration on the main line and the situation that only one kind of terminal units is used.The results verify the scientificalness,feasibility and superiority of the methods in this thesis,and can provide theoretical support for distribution network planning,construction and operation personnel. |
PDF Full Text Request