摘要
短期电力负荷预测对于电力系统的稳定运行和经济调度至关重要,传统预测方法处理大量数据时效率较低。针对短期电力负荷易受外界扰动呈现非线性化、随机的特点,提出一种基于牛顿-拉夫逊向量机(NRBO-SVM)的短期电力负荷预测方法。支持向量机(SVM)通过核函数将历史数据映射到高维特征空间,确定牛顿-拉夫逊算法(Newton-Raphson method)合适的初始解,预计未来电荷趋势,并通过搭建仿真验证。结果表明:NRBO-SVM可较快获得准确初始解,对负荷进行快速预测,且预测精度和收敛速度显著提升。
Short-term power load forecasting is crucial for the stable operation and economic dispatch of the power system,but the traditional forecasting methods are inefficient when dealing with a large amount of data.Aiming at the characteristics of short-term power load,which is susceptible to external disturbances and presents non-linearisation and randomness,a short-term power load forecasting method based on NRBO-SVM is proposed,which maps the historical data into a high-dimensional feature space through kernel function,determines the appropriate initial solution of Newton-Raphson method,anticipates the future charge trend,and verifies the results through simulation construction.The results show that NRBO-SVM can obtain the accurate initial solution faster to predict the load quickly,and the prediction accuracy and convergence speed are significantly improved.
作者
路林艳
徐思文
黄文涛
Lu Linyan;Xu Siwen;Huang Wentao(School of Electrical and Control Engineering,Liaoning University of Engineering and Technology,Huludao Liaoning 125000,China)
出处
《现代工业经济和信息化》
2025年第2期246-247,250,共3页
Modern Industrial Economy and Informationization
关键词
短期电力负荷预测
迭代优化
非线性化
short-term electric load forecasting
iterative optimisation
nonlinearisation
作者简介
第一作者:路林艳(2004-),女,贵州毕节人,辽宁工程技术大学本科在读,研究方向为电力负荷预测。
