苏宜靖,谷炜,赵依,董立,蒋琛,于竞哲
SU Yijing,GU Wei,ZHAO Yi,DONG Li,JIANG Chen,YU Jingzhe

• 1:国网浙江省电力有限公司
• 2:长安大学电子与控制工程学院
• 3:浙江大学电气工程学院

摘要(Abstract)：

综合考虑气象因子对梅雨期负荷进行精确预测,能够为区域电网发电控制、调度安全和经济运行提供日前决策建议。分别构建了基于BP及Elman神经网络的短期负荷模型,引入基础和综合气象因子进行对比研究,发现梅雨期负荷与温度的相关性最强。依据历史气象及负荷数据,训练负荷预测模型,并考虑年度负荷增长趋势对负荷预测结果进行适当修正。结果表明,Elman神经网络在考虑基础或综合气象因子的情况下,对梅雨期日负荷及电量均具有良好的预测特性。
Accurate load forecasting in plum rains considering meteorological factors can provide day-ahead decision-making suggestions for power generation control, dispatching safety and economic operation of regional power grid. Two load forecasting models based on BP and Elman neural network are constructed. The basic and comprehensive meteorological factors are introduced to make a comparative study. It is found that the correlation between temperature and load is the strongest during plum rains. According to the historical meteorological and load data, the load forecasting model is trained, and the load forecasting results are corrected appropriately considering the annual load growth trend. The results show that Elman neural network method has good forecasting characteristics for daily load and electricity quantity in plum rains in consideration of basic or comprehensive meteorological factors.

关键词(KeyWords)： 气象因子;负荷预测;BP神经网络;Elman神经网络;相关性;误差分析
meteorological factors;load forecasting;BP neural network;Elman neural network;relativity;error analysis

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 苏宜靖,谷炜,赵依,董立,蒋琛,于竞哲
SU Yijing,GU Wei,ZHAO Yi,DONG Li,JIANG Chen,YU Jingzhe

DOI: 10.19585/j.zjdl.201912001

参考文献(References)：

• [1]焦润海,苏辰隽,林碧英,等.基于气象信息因素修正的灰色短期负荷预测模型[J].电网技术,2013,37(3):720-725.
• [2]李龙,魏靖,黎灿兵,等.基于人工神经网络的负荷模型预测[J].电工技术学报,2015,30(8):225-230.
• [3]张淑清,任爽,师荣艳,等.基于多变量气象因子的LMBP电力日负荷预测[J].仪器仪表学报,2015,36(7):1646-1652.
• [4]贾逸伦,龚庆武,占劲松,等.基于有效度和相对熵的短期电力负荷组合预测方法[J].电力科学与技术学报,2016,31(2):3-9.
• [5]李博,门德月,严亚勤,等.基于数值天气预报的母线负荷预测[J].电力系统自动化,2015,39(1):137-140.
• [6]刘旭,罗滇生,姚建刚,等.基于负荷分解和实时气象因素的短期负荷预测[J].电网技术,2009,33(12):110-117.
• [7]郇嘉嘉,左郑敏.广东电网负荷特性典型日选取研究[J].电力科学与技术学报,2017,32(1):164-170.
• [8]李伟,王冰,陈献慧,等.基于气象因子权重相似日的短期光伏功率预测[J].广东电力,2018,31(4):59-64.
• [9]史常凯,闫文棋,张筱慧,等.基于BP网络和灰色模型的春节配变重过载预测[J].电力科学与技术学报,2016,31(3):140-145.
• [10]樊唯钦,张伟,李勇钢,等.基于改进人体舒适指数的微电网超短期负荷预测[J].广东电力,2017,30(4):137-142.
• [11]曾燕飞.电力系统短期负荷预测模型设计[J].广东电力,2011,24(4):49-53.
• [12]刘荣,方鸽飞.改进Elman神经网络的综合气象短期负荷预测[J].电力系统保护与控制,2012,40(22):113-117.
• [13]XIA C H,YANG Z H,LI H J.Electric load forecasting using virtual instrument based on dynamic recurrent Elman neural network[C]//2012 Power Engineering and Automation Conference.New York:IEEE,2013:1-4.
• [14]WI Y M,J00 S K,SONG K B.Holiday Load Forecasting Using Fuzzy Polynomial Regression With Weather Feature Selection and Adjustment[J].IEEE Transactions on Power Systems,2012,27(2):596-603.
• [15]晁颖,金烨,朱晶亮,等.考虑光伏发电出力不确定性的年度最大负荷概率预测[J].广东电力,2018,31(9):83-89.