An associative classification method for the operation rule extracting based on decision tree
Xiongwei Hu, Fang Shi, Zhihong Yu, Yanhao Huang, Guangming Lu · 2016
Along with continuous accumulation of power grid data, some big data analysis methods are used to make a judgment on the stability of power grid. However, most of them only concern about whether it is stability or not, lacking of the judgement on stability margin. And some proposed rules are difficult to understand. An improved decision tree (DT) algorithm is used in this paper to find correlative relationship between power flow state and the stability margin of system when specified faults occur and to get some comprehensible rules. Some unrelated features are eliminated through the operation experience, the principal component analysis (PCA) is used to reduce the dimension of features to get the key features which can describe the state of system. All numerical simulations are based on the changes of IEEE 39-bus node test system. The results show the rationality and feasibility of the proposed method.