基于梯度提升决策树静态电压稳定裕度评估

Static voltage stability margin evolution based on gradient boosting regression tree

随着特高压直流输电的发展,电网负荷集中地区无功缺乏的问题日益突出,系统电压稳定与无功功率密切相关,有必要研究电压稳定裕度的在线监测问题,以期为无功补偿装置的协调配合提供参考。基于PMU提供的数据资源,提出一种基于梯度提升决策树的电压稳定裕度在线监测方法,该方法分为离线运行点数据库的建立以及实时评估两部分。通过模拟负荷的变化,求解最优潮流模型,得到各个发电机出力状态以及对应的电压稳定裕度从而形成离线运行点数据库,并在线下对梯度提升决策树进行训练,在线做出电压稳定裕度的评估。实验结果表明,所提方法对裕度的预测准确度高于基于支持向量机回归模型,且时间可以满足实时性的要求。在测量不确定性场景下,当信噪比大于40 dB时,所提方法应对噪声干扰具有较强的鲁棒性。并且基于梯度提升决策树的特征选择能力,可以为PMU的布点提供一定依据,从而实现系统可靠性和经济性的平衡。

With the development of UHVDC transmission,the problem of reactive power shortage in load concentration areas is becoming prominent.The system voltage stability is closely related to reactive power.It is necessary to study the online monitoring of voltage stability margin in order to provide basis for the coordination of reactive power compensation devices.Based on the data resources provided by PMU,an online monitoring method for voltage stability margin based on gradient boosting regression tree(GBRT)is proposed in this paper.The method is divided into two parts:the establishment of and real-time evaluation.The optimal power flow model is solved by simulating the load change,and the output state of each generator and the corresponding voltage stability margin are obtained,in this way,the offline operation point database is formed,and the GBRT is trained offline,and the evaluation of the voltage stability margin is conducted online.The experimental results show that the proposed method has higher prediction accuracy than support vector machine regression model,and the time can meet the requirements of real-time prediction.Under the measurement of uncertainty,when the signal-to-noise ratio(SNR)is greater than 40 dB,the proposed method is robust to noise interference.And based on the feature selection ability of the GBRT,it can provide a basis for the placement of the PMU,thereby achieving a balance between system reliability and economics.