LSTM-WBLS模型在日降水量预测中的应用
作者:
作者单位:

1.南京信息工程大学自动化学院;2.湖北省公众气象服务中心

中图分类号:

TP183

基金项目:

南方海洋科学与工程广东省实验室(珠海)基金(SML2020SP007),国家自然科学基金(62076136)


Application of improved LSTM-WBLS model in daily precipitation forecast
Author:
Affiliation:

1.School of Automation,Nanjing University of Information Science and Technology;2.Hubei public meteorological service center

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

    日降水量变化呈现显著的非线性特征,对其进行精准预测难度很大。近年来,长短时记忆网络(Long-Short Term Memory, LSTM)在降水量预测中优势明显。然而,LSTM的深层结构造成了其存在过拟合、时滞等缺点,从而影响预测精度。注意到宽度学习系统(Broad Learning System, BLS)是直接计算权重,其无需多次迭代的特点可以帮助解决LSTM存在的缺点。但噪声和异常值对建模依旧有着不良影响。基于此,提出改进的加权宽度学习系统(Weighted Broad Learning System, WBLS)。利用加权惩罚因子约束每个样本,分别为正常和异常样本分配高和低权重来增加和减少它们的影响。本文结合了LSTM和WBLS的优点提出LSTM-WBLS日降水量预测模型。选取湖北省巴东站近20年日降水量实测数据进行实证研究,且考虑气压、气温、湿度、风速和日照等因素。结果表明,与现有的预测模型相比,LSTM-BLS模型在所有评价指标上均预测精度最高。特别地,WBLS模块加入解决了LSTM的时滞性问题。而且在不同时间步长下,新模型预测精度亦表现最佳,证明了其稳定性。在运算效率上,LSTM-WBLS和LSTM相比,并未降低。

    Abstract:

    The change of daily precipitation presents significant non-linear characteristics, and it is very difficult to accurately predict it. In recent years, Long-Short Term Memory (LSTM) has obvious advantages in precipitation prediction. However, the deep structure of LSTM causes its shortcomings such as over-fitting and time lag, which affect the prediction accuracy. Note that the Broad Learning System (BLS) directly calculates weights, and its feature of not requiring multiple iterations can help solve the shortcomings of LSTM. However, noise and outliers still have an adverse effect on modeling. Based on this, an improved Weighted Broad Learning System (WBLS) is proposed. Use a weighted penalty factor to constrain each sample, and assign high and low weights to normal and abnormal samples to increase and decrease their influence. This paper combines the advantages of LSTM and WBLS to propose a LSTM-WBLS daily precipitation prediction model. Empirical research is carried out on the actual data of daily precipitation at Badong Station in Hubei Province in the past 20 years, and factors such as air pressure, temperature, humidity, wind speed and sunshine are taken into consideration. The results show that, compared with the existing prediction models, the LSTM-BLS model has the highest prediction accuracy in all evaluation indicators. In particular, the WBLS module has been added to solve the time lag problem of LSTM. Moreover, under different time steps, the prediction accuracy of the new model also performed best, proving its stability. In terms of computational efficiency, LSTM-WBLS has not decreased compared with LSTM.

    参考文献
    [1] 王海鹏, 张斌, 刘祖涵, 等. 基于混沌理论的武汉、宜昌近60年来月降水特征的对比研究[J]. 自然灾害学报, 2012, 21(06): 111-118.
    [2] Wang H P, Zhang B, Liu Z H, et al. Chaos theory-based comparative study on monthly rainfall characteristics in Wuhan and Yichang during recent 60 years[J]. Journal of Natural Disasters, 2012, 21(06): 111-118.
    [3] 张改红. 基于ARIMA模型的渭南市降水量趋势分析与预测[J]. 价值工程, 2019, 38(34): 197-199.
    [4] Zhang G H. Analysis and Prediction of Precipitation Trend in Weinan City Based on ARIMA Model[J]. Value Engineering, 价值工程, 2019, 38(34): 197-199.
    [5] 胡盈,吴静. 基于ARIMA模型的降水空间特征分析及预测[J]. 江西科学, 2021, 39(01): 99-104.
    [6] Hu J, Wu Y. Analysis and Prediction of Precipitation Spatial Characteristics Based on ARIMA Model[J]. Jiangxi Science, 2021, 39(01): 99-104.
    [7] Hartigan J, MacNamara S, Leslie L, et al. Attribution and prediction of precipitation and temperature trends within the Sydney catchment using machine learning[J]. Climate, 2020, 8(10): 120-144.
    [8] Xiang B, Zeng C F, Dong X N, et al. The application of a decision tree and stochastic forest model in summer precipitation prediction in Chongqing[J]. Atmosphere, 2020, 11(5): 508.
    [9] Peng Y Z , Zhao H S, Zhang H, et al. An extreme learning machine and gene expression programming-based hybrid model for daily precipitation prediction[J]. International Journal of Computational Intelligence Systems, 2019, 12(2): 1512-1525.
    [10] 勾志竟, 任建玲, 徐梅, 等. 基于Hadoop的GA-BP算法在降水预测中的应用[J]. 计算机系统应用, 2019, 28(09): 140-146.
    [11] Gou Zhi Jin, Ren Jian Lin, Xu Mei, et al. Application of GA-BP Algorithm Based on Hadoop in Precipitation Forecast[J]. Computer Systems Applications 2019, 28(09): 140-146.
    [12] Mehdi G, Seyyed J, Arash M. Precipitation forecasting by large-scale climate indices and machine learning techniques[J]. Journal of Arid Land, 2020, 12(05): 854-864.
    [13] Shen Chao Peng. A transdisciplinary review of deep learning research and its relevance for water resources scientists[J]. Water Resources Research, 2018, 54(11): 8558–8593.
    [14] 王子岳, 邵曦. 基于S-LSTM模型利用‘槽值门’机制的说话人意图识别[J]. 南京信息工程大学学报(自然科学版), 2019, 11(06): 751-756.
    [15] Wang Zi Yue, Shao Xi. Speaker intention recognition based on S-LSTM model and slot-gate[J]. Journal of Nanjing University of Information Science Technology(Natural Science Edition), 2019, 11(06): 751-756.
    [16] 王朋, 孙永辉, 翟苏巍, 等. 基于小波长短期记忆网络的风电功率超短期概率预测[J]. 南京信息工程大学学报(自然科学版), 2019, 11(04): 460-466.
    [17] Wang Pen, Sun Yong Hui, Zhai Su Wei, et al. Ultra-short-term probability prediction of wind power based on wavelet decomposition and long short-term memory network[J]. Journal of Nanjing University of Information Science Technology(Natural Science Edition), 2019, 11(04): 460-466.
    [18] 罗嘉, 王乐豪, 涂姗姗, 等. 基于LSTM-BLS的突发气象灾害事件中公众情感倾向分析[J/OL]. 南京信息工程大学学报(自然科学版): 1-13[2021-10-07]. http://kns.cnki.net/kcms/detail/32.1801.N.20210628.1426.002.html.
    [19] Luo Jia, Wang Le Hao, Tu Shan Shan, et al.Analysis of public sentiment tendency in sudden meteorological disasters based on LSTM-BLS[J]. Journal of Nanjing University of Information Science Technology(Natural Science Edition), 1-13[2021-10-07]. http://kns.cnki.net/kcms/detail/32.1801.N.20210628.1426.002.html.
    [20] Nguyen D H and Kim J and Bae D. Improving Radar-Based Rainfall Forecasts by Long Short-Term Memory Network in Urban Basins[J]. Water, 2021, 13(6): 776-776.
    [21] 沈皓俊, 罗勇, 赵宗慈. 基于LSTM网络的中国夏季降水预测研究[J]. 气候变化研究进展, 2020, 16(03): 263-275.
    [22] Shen Hao Jun, Luo Yong, Zhao Zong Ci. Prediction of summer precipitation in China based on LSTM network[J]. Climate Change Research, 2020, 16(03): 263-275.
    [23] Ni Lingling, Wang Dong, Vijay P, et al. Streamflow and rainfall forecasting by two long short-term memory-based models[J]. Journal of Hydrology, 2020, 583: 124296-124296.
    [24] Kang Jinle, Wang Huimin, Yuan Feifei, et al. Prediction of precipitation based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China[J]. Atmosphere, 2020, 11(3): 246.
    [25] Chu F, Liang T, Chen L L P, et al. Weighted Broad Learning System and Its Application in Nonlinear Industrial Process Modeling.[J]. IEEE transactions on neural networks and learning systems, 2020, 31(8) : 3017-3031.
    [26] Sepp H, Jü R S. Long Short-Term Memory[J].Neural Computation, 1997, 9(8): 1735-1780.
    [27] Chen C L P , Zhu L, Liu, S F .Universal approximation capability of Broad Learning System and its structural variations[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(4): 1191-1204.
    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

韩 莹,谈昊然,曹允重,罗 嘉. LSTM-WBLS模型在日降水量预测中的应用[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:243
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2021-10-18
  • 最后修改日期:2021-12-07
  • 录用日期:2022-11-09

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