Feasibility Study Regarding the Use of a Conformer Model for Rainfall-Runoff Modeling

被引:3
作者
Lo, WeiCheng [1 ]
Wang, Wei-Jin [2 ]
Chen, Hsin-Yu [1 ]
Lee, Jhe-Wei [1 ]
Vojinovic, Zoran [3 ]
机构
[1] Natl Cheng Kung Univ, Dept Hydraul & Ocean Engn, Tainan 70101, Taiwan
[2] Taiwan Semicond Mfg Co, Hsinchu 300096, Taiwan
[3] UNESCO IHE Delft Inst Water Educ, NL-2601 DA Delft, Netherlands
关键词
CNN; transformer; conformer; rainfall-runoff modeling; RECOGNITION;
D O I
10.3390/w16213125
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Flood disasters often result in significant losses of life and property, making them among the most devastating natural hazards. Therefore, reliable and accurate water level forecasting is critically important. Rainfall-runoff modeling, which is a complex and nonlinear time series process, plays a key role in this endeavor. Numerous studies have demonstrated that data-driven methods, particularly deep learning approaches such as convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and transformers, have shown promising performance in water level prediction tasks. This study introduces the Conformer, a novel deep learning architecture that integrates the strengths of CNNs and transformers for rainfall-runoff modeling. The framework uses self-attention mechanisms combined with convolutional computations to extract essential features-such as water levels, precipitation, and meteorological data-from multiple stations, which are then aggregated to predict subsequent water level series. This study utilized data spanning from 1 April 2006 to 25 July 2021, totaling 5595 days (134,280 h), which were divided into training, validation, and test sets in an 8:1:1 ratio to train the model, adjust parameters, and evaluate performance, respectively. The effectiveness and feasibility of the proposed model are evaluated in the Lanyang River Basin, with a focus on predicting 7-day-ahead water levels. The results obtained from ablation experiments indicate that convolutional computations significantly enhance the ability of the model to capture the local relationships between water levels and other parameters. Additionally, performing convolution computations after executing self-attention operations yields even better results. Compared with other models in simulations, the Conformer model markedly outperforms the CNN, LSTM, and traditional transformer models in terms of the coefficient of determination (R2) and Nash-Sutcliffe efficiency (NSE) indicators. These findings highlight the potential of the Conformer model to replace the commonly used deep learning methods in the field of hydrology.
引用
收藏
页数:22
相关论文
共 50 条
  • [21] Impact of training data size on the LSTM performances for rainfall-runoff modeling
    Boulmaiz, T.
    Guermoui, M.
    Boutaghane, H.
    MODELING EARTH SYSTEMS AND ENVIRONMENT, 2020, 6 (04) : 2153 - 2164
  • [22] A new hybrid artificial neural networks for rainfall-runoff process modeling
    Asadi, Shahrokh
    Shahrabi, Jamal
    Abbaszadeh, Peyman
    Tabanmehr, Shabnam
    NEUROCOMPUTING, 2013, 121 : 470 - 480
  • [23] Soft Computing Techniques for Rainfall-Runoff Modeling and Analysis in River Basin
    Mishra, Pradeep Kumar
    Dwivedi, Rashmi
    WATER RESOURCES MANAGEMENT, 2025,
  • [24] Integration of artificial neural networks with conceptual models in rainfall-runoff modeling
    Chen, JY
    Adams, BJ
    JOURNAL OF HYDROLOGY, 2006, 318 (1-4) : 232 - 249
  • [25] Conjunction of emotional ANN (EANN) and wavelet transform for rainfall-runoff modeling
    Sharghi, Elnaz
    Nourani, Vahid
    Molajou, Amir
    Najafi, Hessam
    JOURNAL OF HYDROINFORMATICS, 2019, 21 (01) : 136 - 152
  • [26] Hybrid Wavelet Neuro-Fuzzy Approach for Rainfall-Runoff Modeling
    Shoaib, Muhammad
    Shamseldin, Asaad Y.
    Melville, Bruce W.
    Khan, Mudasser Muneer
    JOURNAL OF COMPUTING IN CIVIL ENGINEERING, 2016, 30 (01)
  • [27] Rainfall-Runoff Modeling: Comparison of Two Approaches with Different Data Requirements
    A. Bhadra
    A. Bandyopadhyay
    R. Singh
    N. S. Raghuwanshi
    Water Resources Management, 2010, 24 : 37 - 62
  • [28] Rainfall-Runoff Modeling: Comparison of Two Approaches with Different Data Requirements
    Bhadra, A.
    Bandyopadhyay, A.
    Singh, R.
    Raghuwanshi, N. S.
    WATER RESOURCES MANAGEMENT, 2010, 24 (01) : 37 - 62
  • [29] A Comparative Evaluation of the Use of Artificial Neural Networks for Modeling the Rainfall-Runoff Relationship in Water Resources Management
    Turhan, Evren
    JOURNAL OF ECOLOGICAL ENGINEERING, 2021, 22 (05): : 166 - 178
  • [30] Rainfall-runoff modeling using airGR and airGRteaching: application to a catchment in Northeast Algeria
    Yahiaoui, Salima
    Chibane, Brahim
    Pistre, Severin
    Bentchakal, Malika
    Ali-Rahmani, Salah-Eddine
    MODELING EARTH SYSTEMS AND ENVIRONMENT, 2022, 8 (04) : 4985 - 4996