The origin, spread and restoration of large-scale blackout occurred on Nov. 14, 2006 in UCTE Power Grid are described. According to the related investigation reports the direct causes led to the blackout are clarified; the situation of sectionalized UCTE Power Grid and the problems discovered by the blackout and the lessons from it are analyzed. According to the operational condition of interconnected interregional power grids in China, the lessons to be drawn by domestic power grid operation from the large-scale blackout in UCTE Power Grid are put forward.

The origin, spread and restoration of large-scale blackout occurred on Nov. 14, 2006 in UCTE Power Grid are described. According to the related investigation reports the direct causes led to the blackout are clarified; the situation of sectionalized UCTE Power Grid and the problems discovered by the blackout and the lessons from it are analyzed. According to the operational condition of interconnected interregional power grids in China, the lessons to be drawn by domestic power grid operation from the large-scale blackout in UCTE Power Grid are put forward.

The load recovery and the connection of subsystems with power grid during the blackstart of subsystem are researched. Firstly, by means of integrating the rectangular coordinates based weighted optimal multiplier Newton-Raphson method with sensitivity analysis and adopting continuous linear programming method, both mathematical model and algorithm for restoration control to solve the infeasibility of N-1 power flows and out-of-limit states, which appear at load recovery stage, are established; secondly, the models and algorithms to simulate the frequency variation of subsystems at load recovery stage are established; finally, for the problem of connecting subsystems with power grid, a continuous linear programming based optimal generation re-dispatching method is also developed to control the voltage difference between different islands, thus the optimal voltage difference control at the position where the connection is taken place is effectively implemented. The effectiveness of the proposed model and algorithm is verified by simulation results of standard testing system and actual power grid.

The load recovery and the connection of subsystems with power grid during the blackstart of subsystem are researched. Firstly, by means of integrating the rectangular coordinates based weighted optimal multiplier Newton-Raphson method with sensitivity analysis and adopting continuous linear programming method, both mathematical model and algorithm for restoration control to solve the infeasibility of N-1 power flows and out-of-limit states, which appear at load recovery stage, are established; secondly, the models and algorithms to simulate the frequency variation of subsystems at load recovery stage are established; finally, for the problem of connecting subsystems with power grid, a continuous linear programming based optimal generation re-dispatching method is also developed to control the voltage difference between different islands, thus the optimal voltage difference control at the position where the connection is taken place is effectively implemented. The effectiveness of the proposed model and algorithm is verified by simulation results of standard testing system and actual power grid.

The ultimate goal of system restoration after blackout is to restore load fully and rapidly. One of the most important things in load recovery is to control system frequency, and the effective method to keep the frequency stability is to control the balance between load recovery speed and the generator’s responses. A method using genetic simulated annealing algorithms is proposed for determination of load restoration plans on the basis of network reconfiguration, in which system frequency is calculated by extended power flow analysis and the various system constraints are treated by penalty functions. Through the calculation of the genetic group fitness, the optimal recovery path and the maximum load step can be determined. The efficiency and effectiveness of the proposed method are verified by the numerical results on the IEEE 30 and IEEE 118 systems.

The ultimate goal of system restoration after blackout is to restore load fully and rapidly. One of the most important things in load recovery is to control system frequency, and the effective method to keep the frequency stability is to control the balance between load recovery speed and the generator’s responses. A method using genetic simulated annealing algorithms is proposed for determination of load restoration plans on the basis of network reconfiguration, in which system frequency is calculated by extended power flow analysis and the various system constraints are treated by penalty functions. Through the calculation of the genetic group fitness, the optimal recovery path and the maximum load step can be determined. The efficiency and effectiveness of the proposed method are verified by the numerical results on the IEEE 30 and IEEE 118 systems.

Restoring a large-scale power system has always been a complicated and important issue. A lot of research work has been done on different aspects of the whole power system restoration procedure. However, more time will be required to complete the power system restoration process in an actual situation if accurate and real-time system data cannot be obtained. With the development of the wide area monitoring system (WAMS), power system operators are capable of accessing to more accurate data in the restoration stage after a major outage. The ultimate goal of the system restoration is to restore as much load as possible while in the shortest period of time after a blackout, and the restorable load can be estimated by employing WAMS. Moreover, discrete restorable loads are employed considering the limited number of circuit-breaker operations and the practical topology of distribution systems. In this work, a restorable load estimation method is proposed employing WAMS data after the network frame has been reenergized, and WAMS is also employed to monitor the system parameters in case the newly recovered system becomes unstable again. The proposed method has been validated with the New England 39-Bus system and an actual power system in Guangzhou, China.

The optimal load restoration model concerning cold load pickup is formulated as an optimization problem subjected to the system operation constraints including frequency， voltage and generation output， so that the maximum load may be picked up when maintaining reasonable frequency and voltage profiles. With IPLAN programming language provided by PSS/E， particle swarm optimization algorithm is applied to solve the proposed optimization problem. In addition， penalty function is introduced to handle system operation constraints. The proposed method is able to determine the load positions and amounts that can be picked up without disturbing system security and stability. Simulation is performed on a test system and the results demonstrate the effectiveness of the proposed method.

The optimal load restoration model concerning cold load pickup is formulated as an optimization problem subjected to the system operation constraints including frequency， voltage and generation output， so that the maximum load may be picked up when maintaining reasonable frequency and voltage profiles. With IPLAN programming language provided by PSS/E， particle swarm optimization algorithm is applied to solve the proposed optimization problem. In addition， penalty function is introduced to handle system operation constraints. The proposed method is able to determine the load positions and amounts that can be picked up without disturbing system security and stability. Simulation is performed on a test system and the results demonstrate the effectiveness of the proposed method.