Interactive Mechanism Analysis Of Coupled Networks In The Smart Grid And Its Application

A series of issues on developing a stronger smart grid are proposed as the fundamental research for the national strategy. The goals are to operate the grid reliably,safely and economically. However, the expansion and interconnection of grids are increasing the probabilities of outages among areas of higher load as well as the load shedding of those areas. Some research have pointed out the accidents of big blackout are owing to the cascading failures, and the structural and transmission imbalance are also attracting large crowds. The increasing interdependency is not only better to implement the optimal control, but also makes the power grid face the threats from the information systems and also accelerates the spreading of cascading failures among power grids. With the technical development of “interaction”, the increasing penetration of distributed green resource generation and the deregulated power market, the paradigm of distribution systems changes to an active system in which the power customers are becoming individual self-concern objects allowed to inject power to the power grid. In order to improve the energy efficiency, optimize configuration of power systems,and balance the interests of the individuals and the overall market, it needs to study the multi-layered management for the coupled distribution system.This paper is focused on the practical requirements for understanding the complexities on cascading failures, interdependent systems and active distribution system simulation. In order to improve the reliability, the cascading failure mechanism is analyzied as well as the vulnerable indices are proposed. In order to understand the interactions, a cascading failure model on the interdependent power system is proposed, where the impacts from different interdependencies and sturactual characteristics on cascading failures are inverstigated. In order to understand the comprehensive behaviors of self-interested prosumers, a four layer framework is proposed to manange them and guides them to implement some optimal objectives. The main contributions of the thesis are summarized as follows:Firstly, for a better understanding for the mechanism of cascading failures,a weighted topological model for power grids is proposed in which the impedance of transmission lines is chosen as the weighted parameter of each edge.Considering both node and edge differences to define a new structural index named structure entropy which is to quantitatively describe the faults evolution and unbalanced distribution of power network structure. Take consideration of the influences from states and structural characteristics on cascading failures, and thus to identify the vulnerability of power grids with the verification from the domain simulation. The identification method for vulnerable lines consists of the changing of flow entropy, voltage excursion, node-important and factors of geographical position. Simulations on the standard systems have proved our method reliable.Based on the percolation theory, we study the cascading failures considering interactions between power systems and communication systems. We introduce universal interactive model of the coupled system, and explain why the correlation helps to enlarge the blackout. Through improvements of the universal interactive model, considering some topological characteristics of communication networks,the interactive model based on the dynamic power flow calculation is proposed. Taking the IEEE 39-bus system and China’s Guangdong 500-k V system as examples,we study the interdependent influences on blackouts.Furthermore, based on the interactive model between communication networks and power grids, we analyze the impacts on blackout in terms of different interdependencies and structural characteristics. By classifying the structural interdependency, we define the types of the interdependencies and the mathematic model to compare those interdependencies. Taking the IEEE 118-bus system and Changsha power grid as examples, we model the cascading process based on the DC power flow. The simulation results show that the bigger interdependency is,the lower probability of blackout is. Especially,the “degree to degree” of interdependency is better to improve the robustness of power grids against to random failures. There is a threshold value for the inefficient data exchanging which can be also used for predicting the blackout. The double-star network has outstanding structural characteristics to help to delivery outage packages which results in a better control from dispatch centers to relieve cascading failures.Finally,we propose a new scenario of self-sustainable community of electricity prosumers in the emerging distribution system where self-balance is encouraged among prosumers by different balancing premiums.Generation and demand of prosumers are collectively determined by their attitudes,inherent characteristics and prices. Individual attitudes on benefit and comfort are updated after social communications.Accordingly,new roles are given to conventional players with increasing interactions.Load aggregator surrogates prosumers to participate the community market run by a local electricity coordinator who takes responsibilities for operational securities and uses economic incentives to guide prosumers’ physical activities. Regulator is modeled as a policy decision maker to devise overall rules.Multi-agent based simulation with a 4-layered representation is employed to study some features of the community and the best incentive strategies for the desired outcomes of it.