计及光伏及风电并网的电力系统短期负荷预测  

Short-Term Load Forecasting of Power System Considering Photovoltaic and Wind Power Grid Connection

作　　者：姚娟 张晓文 宋嘉[2] 董新伟[2] YAO Juan;ZHANG Xiaowen;SONG Jia;DONG Xinwei(Jiaxing Power Supply Company,State Grid Zhejiang Electric Power Co.,Ltd.,Jiaxing 334201,Zhejiang,China;China University of Mining and Technology,Xuzhou 221000,Jiangsu,China)

机构地区：[1]国网浙江省电力有限公司嘉兴供电公司,浙江嘉兴334201 [2]中国矿业大学,江苏徐州221000

出　　处：《电力大数据》2023年第7期10-22,共13页Power Systems and Big Data

摘　　要：为提升光伏、风电等分布式能源大量接入电网后短期电力负荷的预测精度,促进电网消纳能力提升,本文对光伏出力及短期用电负荷采用小波-径向基函数(RBF)神经网络预测方法。对风力发电首先利用总体平均经验模态分解(EEMD)方法对其功率数据分解;再采用BP神经网络、RBF神经网络、小波神经网络、ELMAN神经网络四种神经网络预测方法进行预测,并用粒子群算法(PSO)和灰色关联度(GRA)修正;最后,利用等效负荷的概念,分析光伏、风力发电并网对于短期电力负荷预测的影响,并将三种模型有效结合,得到了考虑光伏及风力发电并网的电力系统短期负荷预测的等效负荷预测模型。实例分析表明,本文所提方法相较于其他方法在该预测项目上具有相对更高的预测精度。To improve the accuracy of short-term electricity load forecasting after the large-scale integration of distributed energy sources such as photovoltaics and wind power into the grid,and to promote the enhancement of grid accommodation capacitys this paper applies a wavelet-radial basis function(RBF)neural network pr-ediction method to forecast the photovoltaic output and short-term electricity load.For wind power generation,the ensemble mean empirical mode decomposition(EEMD)method is first used to decompose its power data.Then,four neural network prediction methods-back-propagation(BP)neural network,RBF neural network,wavelet neural network,and ELMAN neural network-are employed for prediction,with corrections using particle swarm optimization(PSO)and grey relational analysis(GRA).Finally,using the concept of equivalent load,the impact of the integration of photovoltaic and wind power generation on short-term electricity load forecasting is analyzed.The three models are effectively combined to obtain an equivalent load prediction model for short-term load forecasting in the power system considering the integration of photovoltaic and wind power generation.Case studies demonstrate that the method proposed in this paper exhibits relatively higher prediction accuracy compared to other methods in this forecasting project.

关 键 词：负荷预测 神经网络 模型 等效负荷 相对误差 

分 类 号：TM715[电气工程—电力系统及自动化]