Research On Method Of Topology Optimization And Elasticity Improvement For Power Grids

Smart grid is a man-made complex giant system composed of coupled physical power grid and communication network.It faces security risks from physical powergrid and communication network,such as internal equipment failure,natural disasters,human misoperation and information attacks.In recent years,many blackouts around the world have demonstrated topological vulnerability and insufficient flexibility of smart grid,especially under unpredictable extreme events.On the other hand,the application of new technologies in smart grid,including distributed generations(DGs),flexible transmission,information control and intelligent electrical devices(IEDs),has given smart grid more flexible and effective preventive solution,which can promote the realization and development of elasticity of smart grid.It is therefore urgent to carry out research on topology optimization and elasticity optimization and control for powergrids,which is of great strategic significance for ensuring China's energy security and electricity reliability of critical infrastructure.Based on the idea of intelligent autonomy and self-healing,the quantitative representation theory and measurement method of elasticity,the analysis of topological structure characteristics and the fast detection method of vital nodes and edges,the optimization theory and method of topological elasticity,and the method for system elasticity(including absorption,response and resilience)enhancement are deeply studied by comprehensively using system elasticity theory,complex network theory,optimization and control theory and technologies.This thesis seeks to form a set of theory,method and control technology(covering the whole process of accident handling)for topology optimization and elasticity improvement of powergrids,which would provide fast,accurate and reliable control decision-making and technical support for elasticity of smart grids and enhance the capability of power system to cope with major disasters and emergencies.The performed fundamental research and the related innovative achievement in this thesis are summarized as follows:A quantitative representation theory and measurement method for powergrid elasticity is proposed as follows.First,the concept of powergrid elasticity is defined by mapping powergrids to physical elastic systems,and the connotative characteristics of topology and system elasticity of powergrids are presented.Then,the elasticity-related concepts including external force,stress,strain,elastic coefficient,elastic potential energy and elastic complementary energy are defined and quantified,and the criterion of elastic stability for powergrid is analyzed.Ultimately,from the perspective of topology,a measurement method for topological elasticity of powergrids is proposed by using the total elastic potential energy in the elastic deformation range,and from three dimensions of system function,stress and time,a measurement method of system elasticity(including absorption elasticity,response elasticity and resilience)of powergrid is proposed from the perspective of energy.The proposed theory and methods lay a theoretical foundation for the study about elastic optimization theory and control technology of powergrids.Based on graph theory and complex network theory,the general topology characteristics of powergrid are analyzed deeply.After that,this thesis focuses on the analysis of community(modularization)and hierarchical topology features of powergrids,and reveals the self-similarity among communities and hierarchical subnets,the characteristics of layer-core structure and energy transmission distance between different layers and the key role of high-level subnets in bridging communities.At last,a fast calculation method for powergrid betweenness is proposed in the basis of the proposed decomposition model of community and hierarchy,which is theoretically deduced and strictly demonstrated by the lemmas and corollaries defined in this thesis.The simulation results verify the effectiveness and efficiency of the proposed method and its applicability in dynamic online updating and parallel computing.The problem of maximizing the topological elasticity of powergrid against malicious attacks is studied on the basis of the proposed elastic quantitative representation theory and measurement method.A theoretical optimization model for topology elasticity is established by using the complex network theory and the propagation mechanism of powergrid failures,and the way to maximize the topological elasticity is found through theoretical deduction and demonstration.Furthermore,a topological elasticity optimization algorithm based on a posteriori edge addition(EA)is proposed to maximize the topological elasticity of powergrid.The simulation results show that the proposed method can significantly improve the topological elasticity of powergrids while maintaining their topological functionality.This approach helps to unveil hitherto hidden functions of some inconspicuous components,which in turn,can be used to guide the design of resilient systems,offer an effective and efficient approach for mitigating malicious attacks,and furnish self-healing to reconstruct failed infrastructure systems.A method of absorptive elasticity improvement based on peer-to-peer network protection is presented to minimize the impact of disturbance by means of communication network.The main and backup differential rings are defined at first,and a fault location mechanism is proposed by using the principle of current differential protection and adjusting the threshold of starting current to avoid load current.After that,a fault detection method is proposed for devices.Furthermore,an integrated backup protection strategy based on the presented detection method is proposed to predict the failure status of main protection.This strategy can accelerate the backup location and isolation of electrical faults and minimize the fault isolation range by closing the device-fault-related main protection before electrical fault and starting the backup protection immediately after electrical fault.The effectiveness and engineering practicability of the absorption elasticity improvement method are validated by simulation studies and dynamic experiments.A resilience improvement method based on self-healing control of distributed multi-agent is proposed.A novel reduction model and self-healing model are firstly constructed to reduce the elasticity calculation dimensions and information iteration times.Service restoration is formulated as a multi-objective optimization problem,which is addressed by the constructed network flow model and the proposed service restoration strategy.The strategy combined by the network reconfiguration and intentional islanding algorithms,along with parameter justification,can maximize out-of-service loads restoration with minimum switching times and significantly mitigate variation of loads and intermittence of DGs.Moreover,according to agents' identity attribute,a unified programming framework is constructed,by which each agent can autonomously execute the corresponding tasks according to their own identity attributes and fault locations,the whole task of service restoration is ultimately accomplished through cooperation of agents.From the test results,it can be found that the proposed method can significantly enhance the resilient performance of distribution networks,and be applied to the real distribution networks.