考虑数据缺失的电力系统暂态稳定自适应集成评估方法

Adaptive Integrated Assessment Method for Transient Stability of Power System Considering PMU Data Missing

随着电力物联网概念的提出,暂态稳定评估在电力系统规划运行中扮演着越来越重要的角色。由于同步相量测量单元(PMU)的广泛配置,基于机器学习和PMU在线量测数据的暂态稳定实时评估方法展现出了巨大的发展潜力。针对这类方法在应用中可能因PMU失效而严重影响精度的问题,文中提出了一种考虑数据缺失的电力系统暂态稳定自适应集成评估方法。首先,在保证全网节点可观性的基础上构建考虑PMU重要性的PMU子集集合搜索算法。然后,根据PMU子集对应的特征集训练暂态稳定评估子模型。最后,在任意可能的PMU失效情况下采用自适应加权融合机制构建集成暂态稳定评估模型。在新英格兰10机39节点电力系统上的仿真表明,文中提出的方法在PMU失效造成的数据缺失下仍然能够准确、可靠地进行暂态稳定评估,在鲁棒性、计算量及准确率上相比已有的方法均具有较大优势。

As the concept of power Internet of Things is proposed,transient stability assessment(TSA)plays an important role in planning and operation of power systems.Due to the wide deployment of synchrophasor measurement unit(PMU),the real-time assessment method based on machine learning and online measurement of PMU data has shown great development potential.Aiming at the problem that PMU failure may seriously affect the accuracy of this kinds of method in application,an adaptive integrated assessment method for transient stability of power systems considering data missing is proposed.First,a PMU subset search algorithm considering the importance of PMU is constructed on the basis of guaranteeing the observability of all buses in the power system.Then,the sub-models of TSA are trained according to the corresponding feature sets of the PMU subsets.Finally,an adaptive weighted fusion mechanism is used to integrate a TSA model under any possible PMU failure.The simulation results on the New England 10-machine 39-bus power system show that the proposed model can accurately and reliably assess the transient stability with the data missing caused by PMU failure.Compared with the existing methods,the proposed method has advantages in robustness,computational complexity and accuracy.