To realize the accurate prediction of winter wheat field water demand, based on the measured meteorological data of the experimental field in Yuanyang City, Henan Province, the multi-head attention mechanism was adopted to effectively obtain the relational features between water demand time series according to the size of the correlation coefficients between the meteorological factors and the crop water demand. The 11-, 7- and 4-factor fused multi-head attention (MHA) convolutional-long-short-term memory network (CNN-LSTM) model for predicting winter wheat water demand and comparing with the actual values. The results showed that: 1) The coefficient of determination (R2) between the predicted and measured values of crop water requirement was 0.914, the root mean square error (RMSE) was 0.627 mm, and the relative analysis error (RPD) was 4.243. The R2 was higher and RMSE was lower than that of the 7-factor and 4-factor prediction models. The accuracy of this prediction model increased with more inputs of meteorological factors related to crop water demand. The accuracy of the 4-factor prediction model was the lowest, and the R2 between the predicted value and the measured value of crop water requirement was 0.825, RMSE was 0.946 mm, and RPD was 3.124. In the absence of meteorological data, only the minimum temperature, maximum temperature, average temperature and atmospheric pressure with the highest correlation with the reference crop water requirement could be used to predict the crop water requirement. The number of sensors was reduced and the universality was improved compared with that of 11 factors.2 ) Compared with the classical machine learning model BP, the recurrent neural network model LSTM and the LSTM-Attention with attention mechanism, the CNN-LSTM-MHA model with multi-head attention mechanism has better parameters such as the R2, RMSE and RPD, and the prediction effect is closer to the actual situation of field production. In summary, the CNN-LSTM-MHA crop water requirement prediction model based on the fusion of 11 meteorological factors related to crop water requirement is established, which can improve the problem of insufficient internal feature extraction between data in the convolutional layer. The model can effectively improve the prediction accuracy of winter wheat water requirement and can be used for irrigation decision-making of winter wheat in the field.
采用Penman-Monteith法,此方法充分考虑了理论基础的严谨性,其应用范围极为广泛,能够覆盖各类农业生产环境。更重要的是它在估算参考作物需水量方面拥有极高的精确度,确保了数据分析的可靠性和有效性[17]。联合国粮食及农业组织(Food and Agriculture Organization of the United Nations,FAO) Penman-Monteith修正公式表达如下:
ZhengZ, WangZ K, CaiH J. Estimation of evapotranspiration of winter wheat under deficient irrigation based on SIMDualkc model[J]. Journal of Drainage and Irrigation Machinery Engineering, 2020, 38(2): 212-216 (in Chinese)
[5]
FongBryant N., RebaMichele L., TeagueTina G., BenjaminR .K. Runkle, SuvočarevKosana. Eddy covariance measurements of carbon dioxide and water fluxes in US mid south cotton production[J/OL]. Agriculture, Ecosystems & Environment,(--). DOI:10.1016/j.agee.2019.106813
WangY, ZhangX L, ShiJ L, ShenY J. Climate change and its effect on winter wheat yield in the main winter wheat production areas of China[J]. Chinese Journal of Eco Agriculture, 2022, 30(5): 723-734 (in Chinese)
WangS S, FeiL J, GaoC C, WangX, SunJ S. Experiment on single crop cofficient of summer maize under different furrow irrigations [J]. Transactions of the Chinese Society for Agricultural Machinery,2013,44 (11): 105-111 (in Chinese)
MengW S, SunX H, GuoX H, MaJ J, MaW Y, ZhaoW Y, ZhangW X. Prediction of reference crop water requirement forecast based on artificial bee colony radial basis neural network of artificial bee colony[J]. Water Saving Irrigation,2020(1): 79-83 (in Chinese)
[12]
SidhuR K, KumarR, RanaP S. Long short-term memory neural network-based multi-level model for smart irrigation [J/OL].Modern Physics Letters B, (--). DOI: 10.1142/S0217984920504187
[13]
KararM E, Al-RasheedM, Al-RasheedA. IoT and neural network-based water pumping control system for smart irrigation[J]. Information Sciences Letters, 2020,9(2):107-112
[14]
贾红军.基于计算机控制的智能灌溉系统优化研究[J].农机化研究,2024,46(10):37-42
[15]
JiaH J. The optimization study for intelligent irrigation system based on computer control [J]. Journal of Agricultural Mechanization Research, 2024,46 (10) : 37-42 (in Chinese)
LiZ X, LaiZ Q, LongY M. Prediction of water demand for reference crops based on GA-Elman neural network model[J].Water-Saving Irrigation, 2019(2): 117-120 (in Chinese)
[20]
MaY Z, LvB, WangY F, ShiC Y. Crop water requirement prediction method based on EEMD-Attention-LSTM model [J].Journal of Physics:Conference Serie,2023,2637(1):12-28
ShangZ G, DuanX H. Predicting crop water requirements based on particle swarm optimization and least square support vector machine[J].Computer and Modernization,2018(10):44-47 (in Chinese)
[23]
任 欢,王旭光 .注意力机制综述[J].计算机应用,2021,41(1): 1-6
[24]
RenH, WangX G. Review of attention mechanism[J].Journal of Computer Applications,2021, 41(1): 1-6 (in Chinese)
WuX C, FengX, JiangH. Prediction of carbon emissions based on multi-head attention CNN-LSTM [J/OL].Journal of Chongqing Technology and Business University:Natural Science Edition, 1-11 [--] in Chinese)
DongN, LvX, HouZ A, WangH J, ChenJ, ZhangZ, LiX W, ZhangT, LiD D.Study on irrigation amount of cotton field under mulched drip irrigation based on Penman formula [J].Xinjiang Agricultural Sciences, 2012,49 (4): 617-624 (in Chinese)
ZhangM Y, LiL M, WenZ Z, ZhangS F. Design of intelligent irrigation algorithm based on improved Elman neural network and fuzzy control [J]. Foreign Electronic Measurement Technology, 2021, 40(11): 155-160 (in Chinese)
LangX Q. Design and implementation of remote orchard intelligent irrigation system based on ZigBee and GPRS[D]. Taian:Shandong Agricultural University,2011 (in Chinese)
[33]
AllenR G, PereiraL S, RaesD. FAO Irrigation and Drainage Paper No. 56[M]. Rome: Food and Agriculture Organization, 1998
LiY, DengZ, ZhaiG L, ZongJ, CaiJ M, ZhangW Z. Accumulated temperature model for crop coefficient of winter wheat [J]. Water Saving Irrigation,2015(4):36-40 (in Chinese)
WangX, LiuB, ChenZ C, JuT. 基于多源数据和LSTM模型的县域冬小麦估产[J]. Research of Agricultural Modernization, 2023,44(6):1117-1126 (in Chinese)
[38]
WangW H, XieE Z, LiX. Pyramid vision transformer: A versatile backbone for dense prediction without convolutions [C]// 2021 IEEE/CVF International Conference on Computer Vision, Montreal: Institute of Electrical and Electronics Engineers, 2021
FuT D, HanQ X. Design and research of intelligent irrigation decision-making system based on big data[J]. Journal of Agricultural Mechanization Research,2024,46(11):86-89,95 (in Chinese)
[41]
Amogha HegdeM N, NaikM S, ChaitraS N, MadhaviM, RavichandraA. Prediction and analysis of water requirement in automated irrigation system using artificial neural network (ANN) and lora technology [C]. 2021 IEEE International Conference on Distributed Computing, Montreal: Institute of Electrical and Electronics Engineers, 2021
[42]
WangL, MengZ Q, YangL N. Chinese sentiment analysis based on cnn-bilstm model of multi-level and multiscale feature extraction [J].Computer Science, 2023, 50(5):248-254.
[43]
HuangJ Y, LiJ H, OhJ, KangH. LSTM with spatiotemporal attention for IoT-based wireless sensor collected hydrological time-series forecasting [J]. International Journal of Machine Learning and Cybernetics,2023,14(10): 3337-3352