基于Time2Vec的天然气负荷组合预测方法研究
作者:
作者单位:

西安建筑科技大学

基金项目:

国家自然科学(62072363)


Time2Vec-based natural gas load mix forecasting model study
Author:
Affiliation:

1.XI&2.amp;3.#39;4.&5.AN UNIVERSITY OF ARCHITECTURE AND TECHNOLOGY

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • | |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    针对天然气负荷序列的复杂性和非线性,本文提出一种基于Time2Vec的天然气负荷组合预测模型。首先采用皮尔逊相关系数进行相关性分析,提取出相关性强的气象特征;其次,引入时间向量嵌入层Time2vec,将时间序列转换为连续向量空间,提取相应的时间特征,提高了模型对时间序列信息的计算效率。将Time2Vec提取的时间特征、皮尔逊相关系数选取出的气象特征和原始负荷序列输入到长短期记忆网络(LSTM)和时间卷积网络(TCN)中进行负荷预测,充分利用LSTM的长期记忆能力和TCN的局部特征提取能力。最后,将这两个模型通过注意力机制(Attention)组合起来,根据两个模型的重要程度分别赋予不同的权重,得到最终预测结果。实验结果表明,所提出的基于Time2Vec的天然气负荷组合预测模型具有更强的适应性和更高的精度。

    Abstract:

    Aiming at the complexity and nonlinearity of natural gas load sequence, this paper proposes a natural gas load combination forecasting model based on Time2Vec. Firstly, Time2vec, a time vector embedding layer, is introduced to convert the time series into a continuous vector space, which improves the computational efficiency of the model on the time series information; secondly, the Pearson correlation coefficient is used for the correlation analysis, and the meteorological features with strong correlation are extracted as inputs. The temporal features extracted by Time2Vec and the meteorological features selected by Pearson"s correlation coefficient are inputted into the two models of Long Short-Term Memory Network (LSTM) and Temporal Convolutional Network (TCN) for prediction, which make full use of the long term memory capability of LSTM and the local feature extraction capability of TCN. Finally, these two models are combined through Attention mechanism (Attention), and different weights are assigned according to the importance of the two to obtain the final prediction results. The experimental results show that the proposed Time2Vec-based natural gas load combination prediction model has stronger adaptability and higher accuracy.

    参考文献
    [1] 牛东晓,曹树华,赵磊,等.电力负荷预测技术及其应用[M].北京:中国电力出版社,1998.
    [2] Nepal B, Yamaha M, Yokoe A, et al. Electricity load forecasting using clustering and ARIMA model for energy management in buildings[J]. Japan Architectural Review, 2020, 3(1): 62-76.
    [3] Madhukumar M, Sebastian A, Liang X, et al. Regression model-based short-term load forecasting for university campus load[J]. IEEE Access, 2022, 10: 8891-8905.
    [4] 吴迪,马文莉,杨利君.二次指数平滑多目标组合模型电力负荷预测[J].计算机工程与设计,2023,44(08):2541-2547.
    [5] 基于卡尔曼滤波预测节假日逐点增长率的电力系统短期负荷预测[J].武汉大学学报(工学版),2020,53(02):139-144.
    [6] 孙超,吕奇,朱思曈等.基于双层XGBoost算法考虑多特征影响的超短期电力负荷预测[J].高电压技术,2021,47(08):2885-2898.
    [7] Hong Y Y, Chan Y H. Short-term electric load forecasting using particle swarm optimization-based convolutional neural network[J]. Engineering Applications of Artificial Intelligence, 2023, 126: 106773.
    [8] 肖白,赵晓宁,姜卓等.利用模糊信息粒化与支持向量机的空间负荷预测方法[J].电网技术,2021,45(01):251-260.DOI:10.13335/j.1000-3673.pst.2019.1463.
    [9] 那幸仪,贾俊铖,赵晓筠等.基于小波变换和LSTM模型的城市天然气负荷预测[J].计算机应用与软件,2021,38(12):61-66.
    [10] 赵洋,王瀚墨,康丽等.基于时间卷积网络的短期电力负荷预测[J].电工技术学报,2022,37(05):1242-1251.
    [11] 徐岩,向益锋,马天祥.基于粒子群算法优化参数的VMD-GRU短期电力负荷预测模型[J].华北电力大学学报(自然科学版),2023,50(01):38-47.
    [12] Jalalifar R, Delavar M R, Ghaderi S F. SAC-ConvLSTM: A novel spatio-temporal deep learning-based approach for a short term power load forecasting[J]. Expert Systems with Applications, 2024, 237: 121487.
    [13] 毕贵红, 赵鑫, 李璐, 等.双模式分解 CNN-LSTM集成的短期风速预测模型[J]. 太阳能学报, 2023, 44(3): 191-197.
    [14] Ma Z, Mei G. A hybrid attention-based deep learning approach for wind power prediction[J]. Applied Energy, 2022, 323: 119608.
    [15] Mirza A F, Mansoor M, Usman M, et al. Hybrid Inception-embedded deep neural network ResNet for short and medium-term PV-Wind forecasting[J]. Energy Conversion and Management, 2023, 294: 117574.
    [16] Wen Y, Li Z, Wang X, et al. Traffic demand prediction based on spatial-temporal guided multi graph Sandwich-Transformer[J]. Information Sciences, 2023, 643: 119269.
    [17] Parri S, Teeparthi K, Kosana V. A hybrid VMD based contextual feature representation approach for wind speed forecasting[J]. Renewable Energy, 2023: 119391.
    [18] 赵星宇,吴泉军,展晴晴等.基于贝叶斯优化注意力机制LSTNet模型的短期电力负荷预测[J].科学技术与工程,2023,23(15):6465-6472.
    [19] Bian G, Yang Y, Guo X, et al. Natural Gas Load forecasting based on the fusion of VMD-SE and CNN-BiLSTM-Attention[C]//Journal of Physics: Conference Series. IOP Publishing, 2023, 2504(1): 012048.
    [20] 欧阳福莲,王俊,周杭霞.基于改进迁移学习和多尺度CNN-BiLSTM-Attention的短期电力负荷预测方法[J].电力系统保护与控制,2023,51(02):132-140.
    [21] 廖春花,罗潇,谢睿恒等.基于气象因子的长沙市日最大电力负荷预测对比研究[J].自然灾害学报,2023,32(01):183-190.
    相似文献
    引证文献
    引证文献 [0]
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

王可睿,邵必林.基于Time2Vec的天然气负荷组合预测方法研究[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:103
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2024-01-04
  • 最后修改日期:2024-03-11
  • 录用日期:2024-03-12

地址:江苏省南京市宁六路219号    邮编:210044

联系电话:025-58731025    E-mail:nxdxb@nuist.edu.cn

南京信息工程大学学报 ® 2025 版权所有  技术支持:北京勤云科技发展有限公司