摘要
随着可再生能源机组和灵活性负荷的大量接入,新型电力系统的复杂性和不确定性显著加剧,为电力系统运行方式的调整和计算带来巨大挑战。为此,本文提出一种基于极端梯度提升算法(XGBoost)的电力系统运行方式自动调整模型。该方法首先通过多变量核密度估计对电力系统源-荷不确定性进行建模,得到考虑随机变量相关性的多机组/负荷的联合概率分布。其次,采用场景生成和缩减技术得到电力系统典型运行场景,通过模型驱动生成电力系统典型运行方式样本集。最后,设计XGBoost有监督机器学习模型,训练回归分类树学习电力系统运行方式映射关系,实现电力系统运行方式的在线自动调整。算例结果证明了本文所提方法的有效性和精确性。
The large-scale integration of renewable energy generators and flexible loads significantly increases the complexity and uncertainty of the novel power system,which brings great challenges to power system operation mode adjustment.Hence,this paper proposes an automatic power system operation mode adjustment method based on the extreme gradient boosting algorithm(XGBoost).For the source-load uncertainty modeling,the multivariate kernel density estimation method is adopted to obtain the joint probability distribution of multiple sources and loads considering the correlation of stochastic variables.Then,the typical scenarios for power system operation are ob-tained by scenario generation and reduction.A training sample set of typical power system operation modes is further derived via a model-driven approach.Finally,we design an XGBoost based supervised machine learning model to train the regression classification tree to learn the mapping relationship of power system operation modes,thus reali-zing the online automatic adjustment of power system operation mode.Numerical simulation results demonstrate the validity and accuracy of the proposed method.
作者
刘瑶
金吉良
李明乐
李小腾
向异
贾文皓
杜思君
丁涛
LIU Yao;JIN Jiiang;LI Minge;LI Xiaoteng;XIANG Yi;JIA Wenhao;DU Sijun;DING Tao(State Grid Shaanxi Electric Power Research Institute,Xi’an 710100,China;Northwest Branch of State Grid Corporation of China,Xi’an 710048,China;School of Electrical Engineering,Xi’an Jiaotong University,Xi’an 710049,China)
出处
《电工电能新技术》
CSCD
北大核心
2023年第8期69-78,共10页
Advanced Technology of Electrical Engineering and Energy
基金
国网陕西省电力有限公司科技项目(5226KY220004)。
