电力系统中发电机和负荷参数同时辨识的可行性研究

Feasibility studies on simultaneous identification of parameters of generators and loads in power systems

以往电力系统参数辨识都是基于节点数据逐个进行,而且发电机参数和负荷参数也都是分开辨识。近年来广域测量系统(WAMS)逐步得到推广应用,为广域电力系统参数辨识提供了可能性。针对广域电力系统建模的需求,提出了对发电机和负荷的参数进行整体同时辨识的思路。根据参数轨迹灵敏度的相位关系分析发电机和负荷参数同时辨识的可辨识性。根据参数轨迹灵敏度的幅值大小确定参数辨识的难易程度,并由此确定重点辨识参数。利用BPA软件以及蚁群算法优化辨识参数,在测试系统不同故障情况下进行了参数辨识和对比,验证了发电机和负荷同时辨识的可行性。

In the past, parameter identification of power system was based on the node data, and generator parameters and load parameters were identified separately. Recently with the increased application of the wide-area measurement system （WAMS）, the wide-area power system parameter identification has generally become possible. According to the requirement of the wide-area power system modeling, this paper proposes a new thought to identify parameters of the generator and load model at the same time.The identifiability of the generator and load model parameters can be analyzed by the phase＇s relation of the trajectory sensitivities. The difficulty of parameter identification can be analyzed by the magnitude of the trajectory sensitivities and the key parameters can be determined. With BPA software and the ant colony algorithm, the parameters are identified and the results are compared for different faults of test system. The feasibility of simultaneously identifying parameters of the generator and load model is verified.