| The power industry is currently experiencing significant changes, which poses challenges to current methodology of control and operation of power system. A particular challenge is to control and operate the power system in a smart way, leading to the emerging "smart grids" technology. A smart grid is a modernized electrical grid that uses information and communications technology to improve the efficiency, reliability, economics, and susceptibility of the production and distribution of electricity. Smart grids require the upgrade of current power grids to support distributed generation and storage capabilities, to accommodate a greater emphasis on customer participation as well as to meet environmental targets.On one hand, the modern grids are interconnected operating by multiple independent system operators. On the other, the renewable resources and generation are usually distributed geographically. Distributed solutions can share computationally expensive tasks among multiple distributed controllers that work in parallel. A well-designed distributed control solution is able to solve a complex problem much faster. In addition, compared to centralized solutions, distributed solutions is usually less expensive to implement, flexible, and robust against single-point failures. Thus, distributed solutions can improve response speed and reliability, and are promising choice for future power grids control and operation.;To address needs of future smart grids, the author developed distributed supporting algorithms for power system control and operation. Be specific, distributed algorithms for state estimation, active power coordination for doubly-fed introduction generators (DFIGs), optimal active and reactive power control are all investigated.;For data processing, a distributed topology identification and state estimation algorithms are proposed for multi-area data acquisition system that can provide basic data for advanced applications. The proposed algorithms are based on distributed subgradient method that can solve a convex optimization problem in a distributed way while still obtain same solutions as centralized methods. For advanced applications of power systems, one of the important applications is real-time supply-demand balance control. When penetration of renewable generation is high, the available generation may be, more than needed, especially for autonomous operating microgrid. A fully distributed control scheme is presented to coordinate the operations of multiple DFIGs in a microgrid to demonstrate the real-time supply-demand balance control of power systems. By controlling the utilization levels of DFIGs at a common value, the supply-demand balance can be maintained, which ensure stable operation of power grids. For optimization of power systems, a multi-agent system based distributed control solution that can realize online optimal generation control is proposed. The solution is designed based upon an improved distributed gradient algorithm, which can address both equality and inequality constraints. Compared with centralized solutions, the distributed control solution not only can achieve comparable solutions but also can response timely when the power system experiences change of operating conditions. To improve energy efficiency and voltage profiles of the power grids under different operating conditions, a distributed multiple agent system based optimal reactive control solution is presented in our research. The control solution is based on distributed computation of subgradient, and can be implemented in distributed fashion. The proposed solution is able to provide comparable steady state performance as that of centralized optimization solutions.;The effectiveness of all the algorithms proposed has been demonstrated through simulation. In addition, a multi-agent system (MAS) based real-time experimental platform is also developed to test the distributed control solutions. The research work of this dissertation demonstrates that distributed control and operation of power grids is possible, and which is also a promising option for future smart grids. |
