基于孤立森林与随机子空间的小干扰稳定评估

Small signal stability assessment based on isolation forest with stochastic subspace identification

环境激励下的电力系统随机响应蕴含着系统丰富的动态信息,通过随机子空间技术提取出随机响应中的小干扰稳定特征参数,但是提取过程中往往容易混入异常数据,导致无法判断出系统的真实状态。针对此问题提出了基于随机子空间与孤立森林的小干扰稳定评估方法,使用孤立森林算法对随机子空间技术提取的小干扰稳定特征参数结果进行异常值检测,有效改善了随机子空间技术获取系统的特征参数(振荡频率、阻尼比)的准确性。最后对IEEE 16机系统进行实例仿真,仿真结果表明了该方法的可行性。

The stochastic response of power system under environmental excitation contains rich dynamic information of the system,and the small signal stability characteristic parameters in the stochastic response can be extracted by the stochastic subspace technique,but the extraction process is often easy to mix in abnormal data,which leads to the inability to judge the real state of the system.To address this problem,a small signal stability evaluation method based on random subspace and isolation forest is proposed,and the isolation forest algorithm is used to detect outliers in the results of small signal stability characteristic parameters extracted by random subspace technique,which effectively improves the accuracy of obtaining the characteristic parameters(oscillation frequency and damping ratio)of the system.Finally,an example simulation is performed in the IEEE16 machine system,and the simulation results show the feasibility of the method.