基于分类回归树算法的在线静态电压稳定裕度评估

随着电力负荷的不断增加,电压稳定问题日益凸显;为防止电压失稳,电网运行人员迫切需要了解电网当前的运行状态。提出运用分类回归树算法构建决策树模型,对静态电压稳定裕度进行在线评估。首先根据PV曲线分析和电压稳定裕度计算获得决策树训练样本集;然后通过模态分析和相关系数分析降低属性集的维度,确定训练决策树的输入属性;最后利用离线数据,运用决策树分类回归树算法训练出在线电压稳定裕度的评估规则。以某省实际电网为例对所提方法进行验证,其决策规则具有较高的准确性,能够实时地评估电网运行状态。

Online Evaluation on Static Voltage Stability Margin Based on Classification and Regression Tree Algorithm

Along with the continuous growth of power load,the issue of voltage stability has become increasingly impor⁃tant.To avoid voltage instability,system operators need to know the operating condition of one power system in its pres⁃ent state.In this paper,a classification and regression tree(CART)algorithm is proposed to build a decision tree(DT)model for the online evaluation on static voltage stability margin.First,the training sample set of DT is obtained accord⁃ing to the P-V curve analysis and the calculation results of voltage stability margin.Second,the dimension of the attri⁃bute set is reduced through modal analysis and correlation coefficient analysis,so that the input attribute of DT for train⁃ing can be determined.Finally,the CART algorithm is used to train the online evaluation criteria for voltage stability margin.An actual power system in one province of China is taken as an example to verify the proposed method,and re⁃sults show that the criteria of DT have higher accuracy and can evaluate the power system’s operation states in real time.