Research On Fast Transient Stability Simulation

As the scale of an interconnected power system is enlarging increasingly,the structure and operation conditions of power systems are becoming more complicated. In order to assure the security of power system after contingency,the requirement for faster transient stability simulation is also more urgent.It is well known that fast transient stability evaluation method for power system operation is of great significance.Numerical integration method has been used widely and its drawback is that computation speed is too slow,so improving the speed is the main difficulty in time-domain simulation.Though lots of methods were presented so far,none of them is fast and reliable enough for real-time practical use.The main works of this paper focus on how to improve the speed of time-domain simulation effectively.After the research background and different kinds of transient stability evaluation methods introduced,according to the characteristic of solving the network equation in transient analysis,this paper firstly presents a generator parameter equivalent methodology which can be applied in fast transient stability analysis by solving network equation without iteration.This methodology theoretically assures the equation x~″_q=x~″_d met on the basis of equivalent damping windings,then other q-axis parameters can be optimized by particle swarm optimization(PSO) to avoid solving network equation iteratively.One-machine infinite test system demonstrates that optimized parameters with good adaptation can be applied in different operation conditions.Furthermore,a multi-machine system considering another load model i.e. induction motor can also make use of optimized parameters transplanted from one-machine system,before and after generator parameter equivalent the difference of rotor angles is small.Different ways to solving network equation are explained in detail next,in order to prove that the non-iterative method presented before is exact enough,different rotor angles with different step length and different means to solve network equation are also compared particularly.In addition power angle prediction would also reduce computation time in time-domain simulation and the iteration of solving network equation,so angle prediction based on artificial neural networks(ANN) is applied in transient analysis.The test system taking into account different operation conditions shows that generator optimized parameters computed by particle swarm optimization algorithm and power angle prediction can decrease the iteration times of solving network equations effectively and determine whether the system is secure or not. This study explains that optimized parameters and angle prediction are expected to be applied to numerical simulation in various power systems.In the end,the author makes a summary of this paper and points out that the more works need be done in the future.