基于改进SSP抽样技术的发输电系统可靠性评估

Transmission system reliability evaluation based on the improvement of state space partition method of sampling technology

针对传统发输电系统可靠性评估方法抽样次数多、耗时长等问题,提出一种基于分层自寻优抽样技术的状态空间分割方法。通过合理划分全状态空间,选用解析法和模拟法对两个子空间分别进行状态选择,并通过分层自寻优抽样方法改进模拟法的抽样技术,优化分配高重故障状态的抽样次数,减小状态空间分割法的抽样方差。使用可靠性测试系统IEEE-RTS 79以及修改后的测试系统对所提方法进行验证。仿真算例结果分析表明,该方法不仅适用于不同规模的系统评估,而且提升了传统状态空间分割法的计算效率。

In order to solve the problem that the reliability evaluation method of traditional transmission system has a high sampling frequency and a big time consuming, this paper proposes an improved state space partitioning based on automatic optimization stratified cluster sampling method. Through reasonable division of the whole state space, the method selects the analytic method and Monte Carlo Sampling to state choice in two subspaces, and through the automatic optimization stratified cluster sampling method, the method improves the optimal allocation of heavy fault space sampling times of simulation method and reduces the sampling variance of state space partitioning. The IEEE reliability test system is used to test the proposed methodology. Simulation results show that the method is not only suitable for different sizes of system evaluation, but also improves the computational efficiency of traditional state space partition method.