LSTM在输变电设备缺失值填补中的应用

输变电设备是电网的重要组成部分,其状态量值表征了设备的基本运行状态。由于一些不可控因素,在采集时会有一些"空值"。这些缺失值不仅意味着信息空白,更重要的是它会影响后续数据挖掘和统计分析等工作的进行。文中提出了一种基于长短时记忆网络(LSTM)的缺失值填补方法,与经典的数据挖掘方法进行对比,实验表明所提方法的填补结果在均方根误差这一评价指标上有20%的提升。同时还综合考虑了同一设备下其他不同状态量以及气象因素的影响。最后,利用所述方法对国网某省公司电网线路的在线监测数据进行了缺失值填补和验证,结果表明该方法在常规条件下具有较好的填补效果。

Application of LSTM in filling missing value of power transmission and transformation equipment

Power transmission and transformation equipment is an important part of the power grid, and its state value represents the basic operation of the equipment. Due to some uncontrollable factors, there are some null values in the collection. These missing values mean not only information gaps, but more importantly, they will affect the subsequent data mining. In this paper, a method of missing value imputation based on Long Short-Term Memory networks (LSTM) is proposed. Compared with the classical data mining method, the results show that the use of the LSTM-based method has a 20% improvement on the root mean square error. At the same time, we also considers the influence of other different state variables and meteorological factors under the same equipment. Finally, the method is used to fill the missing data in State Grid. The results show that the proposed method can meet the requirement of filling under the normal conditions.