基于时间特征分析的NST-IRN组合模型短期负荷功率预测

万炜兴; 谢丽蓉; 张龙军; 卞一帆; 马兰

doi:10.19912/j.0254-0096.tynxb.2024-1869

PDF(2328 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (7) : 62-72. DOI: 10.19912/j.0254-0096.tynxb.2024-1869

第二十七届中国科协年会学术论文

基于时间特征分析的NST-IRN组合模型短期负荷功率预测

万炜兴¹, 谢丽蓉¹, 张龙军², 卞一帆¹, 马兰¹

作者信息 +

SHORT-TERM LOAD POWER FORECASTING USING NST-IRN COMBINED MODEL BASED ON TEMPORAL FEATURE ANALYSIS

Wan Weixing¹, Xie Lirong¹, Zhang Longjun², Bian Yifan¹, Ma Lan¹

Author information +

文章历史 +

摘要

针对在短期负荷功率预测过程中,因负荷功率波动大而导致单一模型存在预测精度低的问题,提出基于时间特征分析的短期负荷功率NST-IRN组合预测模型。首先,深度挖掘负荷功率的时间特性变化,将其分解为趋势成分以及循环分量,构建新型时间序列（NTS）模型。其次,考虑多时间尺度输入特征与日类型对负荷功率的影响,构建含特征输入结构与深度学习结构的改进残差神经网络（IRN）模型。最后,利用D-S证据理论对NTS与IRN模型的预测结果进行权重融合,以获取最终负荷预测结果。以ISO New England的真实负荷数据进行仿真实验,结果表明所提模型具有较高的预测精度和鲁棒性。

Abstract

In the process of short-term load power forecasting, in view of the large fluctuation of load power, which leads to the low prediction accuracy of a single model, a short-term load power NST-IRN combined forecasting model based on time feature analysis is proposed. Firstly, the temporal characteristics of load variations are thoroughly analyzed, decomposing them into trend components to construct a new time series （NTS） model. Secondly, considering the impact of multi time scale input features and day types on load power, a improve residual neural （IRN） model with feature input structure and deep learning structure is constructed. Finally, the D-S evidence theory is used to weight and fuse the prediction results of the NTS and IRN models to obtain the final load forecasting results. Simulation experiments were conducted using real load data from ISO New England, and the results showed that the proposed model has high prediction accuracy and robustness.

导出引用

万炜兴, 谢丽蓉, 张龙军, 卞一帆, 马兰. 基于时间特征分析的NST-IRN组合模型短期负荷功率预测[J]. 太阳能学报. 2025, 46(7): 62-72 https://doi.org/10.19912/j.0254-0096.tynxb.2024-1869

Wan Weixing, Xie Lirong, Zhang Longjun, Bian Yifan, Ma Lan. SHORT-TERM LOAD POWER FORECASTING USING NST-IRN COMBINED MODEL BASED ON TEMPORAL FEATURE ANALYSIS[J]. Acta Energiae Solaris Sinica. 2025, 46(7): 62-72 https://doi.org/10.19912/j.0254-0096.tynxb.2024-1869

中图分类号： TM715

参考文献

[1] 王印松, 吕率豪. 基于改进时间卷积网络的微电网超短期负荷预测[J]. 太阳能学报, 2024, 45(6): 255-263.
WANG Y S, LYU S H.Ultra-short-term power load prediction of micro-grid based on improved temporal convolutional network[J]. Acta energiae solaris sinica, 2024, 45(6): 255-263.
[2] LIU C, WU W Z, XIE W L, et al.Application of a novel fractional grey prediction model with time power term to predict the electricity consumption of India and China[J]. Chaos, solitons & fractals, 2020, 141: 110429.
[3] DUDEK G, PELKA P, SMYL S.A hybrid residual dilated LSTM and exponential smoothing model for midterm electric load forecasting[J]. IEEE transactions on neural networks and learning systems, 2022, 33(7): 2879-2891.
[4] CHODAKOWSKA E, NAZARKO J, NAZARKO Ł.ARIMA models in electrical load forecasting and their robustness to noise[J]. Energies, 2021, 14(23): 7952.
[5] ZHAO Z Y, ZHANG Y, YANG Y J, et al.Load forecasting via grey model-least squares support vector machine model and spatial-temporal distribution of electric consumption intensity[J]. Energy, 2022, 255: 124468.
[6] DENG C R, ZHANG X Y, HUANG Y M, et al.Equipping seasonal exponential smoothing models with particle swarm optimization algorithm for electricity consumption forecasting[J]. Energies, 2021, 14(13): 4036.
[7] 方娜, 陈浩, 邓心, 等. 基于VMD-ARIMA-DBN的短期电力负荷预测[J]. 电力系统及其自动化学报, 2023, 35(6): 59-65.
FANG N, CHEN H, DENG X, et al.Short-term power load forecasting based on VMD-ARIMA-DBN[J]. Proceedings of the CSU-EPSA, 2023, 35(6): 59-65.
[8] 苏学能, 刘天琪, 曹鸿谦, 等. 基于Hadoop架构的多重分布式BP神经网络的短期负荷预测方法[J]. 中国电机工程学报, 2017, 37(17): 4966-4973, 5216.
SU X N, LIU T Q, CAO H Q, et al.A multiple distributed BP neural networks approach for short-term load forecasting based on Hadoop framework[J]. Proceedings of the CSEE, 2017, 37(17): 4966-4973, 5216.
[9] 骆钊, 吴谕侯, 朱家祥, 等. 基于多尺度时间序列块自编码Transformer神经网络模型的风电超短期功率预测[J]. 电网技术, 2023, 47(9): 3527-3537.
LUO Z, WU Y H, ZHU J X, et al.Wind power forecasting based on multi-scale time series block auto-encoder Transformer neural network model[J]. Power system technology, 2023, 47(9): 3527-3537.
[10] 欧阳静, 杨吕, 尹康, 等. 基于ALIF-LSTM多任务学习的综合能源系统短期负荷预测[J]. 太阳能学报, 2022, 43(9): 499-507.
OUYANG J, YANG L, YIN K, et al.Short-term load forecasting method for integrated energy system based on ALIF-LSTM and multi-task learning[J]. Acta energiae solaris sinica, 2022, 43(9): 499-507.
[11] 许言路, 卢悦, 朱冰, 等. 基于FFT优化ResNet模型的短期负荷预测方法[J]. 控制工程, 2019, 26(6): 1085-1090.
XU Y L, LU Y, ZHU B, et al.Short-term load forecasting method based on FFT optimized ResNet model[J]. Control engineering of China, 2019, 26(6): 1085-1090.
[12] 李国庆, 刘钊, 金国彬, 等. 基于随机分布式嵌入框架及BP神经网络的超短期电力负荷预测[J]. 电网技术, 2020, 44(2): 437-445.
LI G Q, LIU Z, JIN G B, et al.Ultra short-term power load forecasting based on randomly distributive embedded framework and BP neural network[J]. Power system technology, 2020, 44(2): 437-445.
[13] 黄文琦, 梁凌宇, 王鑫, 等. 基于变量选择与Transformer模型的中长期电力负荷预测方法[J]. 浙江大学学报(理学版), 2024, 51(4): 483-491, 500.
HUANG W Q, LIANG L Y, WANG X, et al.Mid-long term power load forecasting based variable selection and Transformer model[J]. Journal of Zhejiang University (science edition), 2024, 51(4): 483-491, 500.
[14] 刘杰, 从兰美, 夏远洋, 等. 基于DBO-VMD和IWOA-BILSTM神经网络组合模型的短期电力负荷预测[J]. 电力系统保护与控制, 2024, 52(8): 123-133.
LIU J, CONG L M, XIA Y Y, et al.Short-term power load prediction based on DBO-VMD and an IWOA-BILSTM neural network combination model[J]. Power system protection and control, 2024, 52(8): 123-133.
[15] HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA, 2016: 770-778.
[16] 樊江川, 于昊正, 刘慧婷, 等. 基于多分支门控残差卷积神经网络的短期电力负荷预测[J]. 中国电力, 2022, 55(11): 155-162, 174.
FAN J C, YU H Z, LIU H T, et al.Short-term load forecasting based on multi-branch residual gated convolution neural network[J]. Electric power, 2022, 55(11): 155-162, 174.
[17] NIELSEN H A, NIELSEN T S, MADSEN H, et al.Optimal combination of wind power forecasts[J]. Wind energy, 2007, 10(5): 471-482.
[18] WANG B W, CHE J F, WANG B, et al.A solar power prediction using support vector machines based on multi-source data fusion[C]//2018 International Conference on Power System Technology (POWERCON), Guangzhou, China, 2018: 4573-4577.
[19] XIA D, WU J Y.Study on the integrated model for short-term wind power forecasting based on D-S evidence theory[C]//2012 Asia-Pacific Power and Energy Engineering Conference. Shanghai, China, 2012: 1-4.
[20] DING N, BÉSANGER Y, WURTZ F. Next-day MV/LV substation load forecaster using time series method[J]. Electric power systems research, 2015, 119: 345-354.
[21] HARDY M A.Regression with dummy variables[M]. sage, 1993: 128-132.
[22] SAID S E, DICKEY D A.Testing for unit roots in autoregressive-moving average models of unknown order[J]. Biometrika, 1984, 71(3): 599-607.
[23] KWIATKOWSKI D, PHILLIPS P C B, SCHMIDT P, et al. Testing the null hypothesis of stationarity against the alternative of a unit root How sure are we that economic time series have a unit root?[J]. Journal of econometrics, 1992, 54(1-3): 159-178.
[24] AKDI Y, GÖLVEREN E, OKKAOĞLU Y. Daily electrical energy consumption: periodicity, harmonic regression method and forecasting[J]. Energy, 2020, 191: 116524.
[25] CHEN K J, CHEN K L, WANG Q, et al.Short-term load forecasting with deep residual networks[J]. IEEE transactions on smart grid, 2019, 10(4): 3943-3952.
[26] KLAMBAUER G, UNTERTHINER T, MAYR A, et al. Self-normalizing neural networks[EB/OL].2017: 1706.02515. https://arxiv.org/abs/1706.02515v5.
[27] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA, 2017: 2261-2269.
[28] ZHAO L M, WANG J D, LI X, et al. On the connection of deep fusion to ensembling[J]. arXiv preprint, arXiv:1611, 07718, 2016.
[29] DU Y W, ZHONG J J.Generalized combination rule for evidential reasoning approach and Dempster-Shafer theory of evidence[J]. Information sciences, 2021, 547: 1201-1232.