| With the continues expansion of power system scale and the development of power grid interconnection, the cascading failure caused by disastrous weather or other accidental factors will be more likely to produce large-scale blackouts. Since2003, several blackouts have occurred in United States, Canada, Russia, Japan and some European countries. There were also some blackouts caused by nature disaster occurred in China in recent years. The existing operation experience and theoretical research shows that, even new equipments and technologies can make power system operate more safe and stable, the blackout is unavoidable. But the outage area and time can be reduced by appropriate prevention strategy, timely emergency control and proper restoration plan.Reasonable restoration subsystem partitioning can increase the restoration efficiency and ensure the reliability. Therefore, it is necessary to analyze the characteristics of power system restoration problem and improve restoration subsystem partition method. Usually, power system restoration is accomplished by three stages:black-start, network reconfiguration and load restoration. The feasibility and efficiency of network reconfiguration depend on the given restoration sequence and path. However, different evaluation methods may lead to different optimal schemes. As there are various attributes used in network reconfiguration evaluation and the evaluation value are expressed in different forms. A comprehensive scheme evaluation method is needed to help operator make reasonable decisions. In addition, the restoration control applied now is usually based on the pre-established restoration plan which is difficult to meet the developing power grid or the actual situation. Therefore, it is needed to develop a power system restoration decision support system to remedy the deficiency of the existing dispatching platform. Based on literature review of researches on power system restoration, thorough studies on the above topics have been made in this dissertation. The main research work and innovation are as follows.1) Based on the community structure in complex network theory, a new dynamic partition method is proposed to solve the subsystem partition problem in power system restoration. The proposed method divides the partition process into agglomerate stage and split stage. The agglomerate stage deals with some special requirements in black start stage and the following split stage finds the tie-line among subsystems. Simulation of Shandong Power Grid shows that the partition scheme formed by this method is feasible, reliable and is comply with the power system operation principle of layering and zoning. Besides, the proposed method can help with the optimal allocation of black start unit.2) A new evaluation method is proposed based on UMADM(Uncertain Multiple Attribute Decision Making) theory in determining the skeleton-network reconfiguration scheme. This method first classifies the attributes into constraint attributes and optimization attributes based on their role in reconfiguration, and then the weights of attributes are determined in different ways according to their type. The interval number is introduced to integrate quantitative and qualitative attributes. By employment of this method, all attributes are taken into account reasonably and objectively, and the result can also reflect the requirements of the subject of evaluation. Simulation of Jinan district shows that the proposed method can consider different attributes comprehensively and help the operator make reasonable decision.3) The design and development strategy of power system restoration decision support system is preliminary studied. Based on its position in power system distribute platform, the main function and data requirement of the system are analyzed and then the related data tables are designed. In addition, by taking network reconfiguration stage as an example, the local optimization method which is applied to generate restoration strategy in the system is illustrated. The research offers reference for the further development of restoration decision support system. |
PDF Full Text Request