基于XGBoost算法的电压暂降预估模型

Voltage Sag Prediction Based on XGBoost Algorithm

由于电力工业的发展,电压暂降现象出现的概率大大增加,对工业生产造成重大损失。在此背景下,电压暂降的预估显得尤为重要。为了实现对电压暂降这一暂态过程的预估,以避免造成较大的经济损失,本文通过研究电压暂降所具有的幅值变化和持续时间,结合决策树等模型提出了一种电压预估方法。本文还研究了基于Boosting算法的提升树XGBoost模型,提出了基于XGBoost模型的电压暂降预估模型,并通过在simulink搭建IEEE9的电网仿真模型进行了数据处理和预估模型验证,分析了模型的可行型,由仿真结果可知,本文所建模型具有一定的创新性与稳定性,具有实际的预估效果。

Due to the development of electric power industry,the probability of voltage sag phenomenon is greatly increased,causing heavy losses to industrial production.In this context,the prediction of voltage sag is particularly important.In order to realize the prediction of voltage sag,a transient process,and avoid large economic losses,this paper proposes a voltage prediction method by studying the amplitude change and duration of voltage sag and combining decision tree and other models.After that,this paper studies the lifting tree XGBoost model based on Boosting algorithm,and proposes a voltage dip prediction model based on XGBoost model,and through building the IEEE9 power grid simulation model in simulink,according to the voltage sag data obtained from the simulation,data processing and prediction model verification were carried out,and the feasibility of the model was analyzed.From the simulation results,it can be seen that the model built in this paper has certain innovation and stability,and has actual prediction effect.