An Efficient Hybrid Model With Harris Hawks Optimization Algorithm for Predicting Oat Water
DOI:
https://doi.org/10.4108/airo.12011Keywords:
Harris Hawk Optimization, TCN-BiLSTM-MHA, oat water demandAbstract
Oats are a cold-tolerant and high-yielding cultivated forage crop, which have relatively high requirements for water management. To improve the forecasting accuracy of irrigation requirements for oats, this paper proposes a novel hybrid neural network architecture, whose parameters are refined using the Harris Hawk optimization algorithm. It directly addresses two prevalent shortcomings in current predictive models: the typically imprecise manual adjustment of hyperparameters, and the inadequate modeling of both spatial and temporal dependencies with the data. By integrating these methodological improvements, the proposed method aims to achieve more precise and robust forecasts. Firstly, the hybrid model integrates a temporal convolutional network, a bidirectional long short-term memory network, and a multi-head attention mechanism. This method leverages multi-head attention to enrich feature representation, thereby facilitating a more comprehensive capture of the temporal dynamics inherent in alpine oat water demand. Secondly, the Harris Hawk optimization method is introduced to optimize the model’s hyperparameters, effectively avoiding local optimum. Experimental results on the oat water demand and environmental dataset from 2019 to 2023 indicate that the hybrid model achieved a mean absolute error and root mean square error of 0.3432 and 0.4863, respectively, thus representing reductions of approximately 52.73% and 46.39%compared to the traditional Long Short-Term Memory (LSTM) model. The coefficient of determination increased by about 15.31%. Ablation study results demonstrate that the complete hybrid model achieved Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of 0.3432 and 0.4863, respectively, which represent reductions of approximately 52.89% and 43.75% compared to the baseline model, the coefficient of determination improving by about 22.16%. Compared with other methods, this method has a distinct advantage in forecasting the precision of oat water demand, which offers technical and decision support for smart irrigation.
Downloads
References
[1] Esmaeeli, A. and Sarrafzadeh, M.H. (2023) Reducing freshwater consumption in pulp and paper industries using pinch analysis and mathematical optimization. Journal of Water Process Engineering 53: 103646.
[2] Xing, Y. and Wang, X. (2024) Precision agriculture and water conservation strategies for sustainable crop production in arid regions. Plants 13(22): 3184.
[3] Hussain, S., Mubeen, M., Nasim, W., Fahad, S., Ali, M., Ehsan, M.A. and Raza, A. (2023) Investigation of irrigation water requirement and evapotranspiration for water resource management in southern punjab, pakistan. Sustainability 15(3): 1768.
[4] Capodaglio, A.G. (2024) Urban water supply sustainability and resilience under climate variability: Innovative paradigms, approaches and technologies. ACS ES&T Water 4(12): 5185–5206.
[5] Moshayedi, A.J., Roy, A.S., Sambo, S.K., Zhong, Y. and Liao, L. (2022) Review on: The service robot mathematical model. EAI Endorsed Transactions on AI and Robotics 1(1): 1–19.
[6] Greff, K., Srivastava, R.K., Koutnik, J., Steunebrink, B.R. and Schmidhuber, J. (2016) Lstm: A search space odyssey. IEEE Transactions on Neural Networks and Learning Systems 28(10): 2222–2232.
[7] Zhang, X., Zhang, Q., Zhang, G., Nie, Z., Gui, Z. and Que, H. (2018) A novel hybrid data-driven model for daily land surface temperature forecasting using long short-term memory neural network based on ensemble empirical mode decomposition. International Journal of Environmental Research and Public Health 15(5): 1032.
[8] Moshayedi, A.J., Roy, A.S., Kolahdooz, A. and Shuxin, Y. (2022) Deep learning application pros and cons over algorithm. EAI Endorsed Transactions on AI and Robotics 1(1): e7.
[9] Li, J., Zhu, D. and Li, C. (2022) Comparative analysis of bpnn, svr, lstm, random forest, and lstm-svr for conditional simulation of non-gaussian measured fluctuating wind pressures. Mechanical Systems and Signal Processing 178: 109285.
[10] Long, X., Wang, J., Gong, S., Li, G. and Ju, H. (2022) Reference evapotranspiration estimation using long short-term memory network and wavelet-coupled long short-term memory network. Irrigation and Drainage 71(4): 855–881.
[11] Wu, C., Lu, H., Liu, D., Wang, C., Gan, J. and Li, Z. (2025) Research on gas pipeline leakage prediction model based on physics-aware gl-translstm. Biomimetics 10(11): 743.
[12] Paidipati, K. and Banik, A. (2020) Forecasting of rice cultivation in india–a comparative analysis with arima and lstm-nn models. EAI Endorsed Transactions on Scalable Information Systems 7(24).
[13] Han, Z., Zhao, J., Leung, H., Ma, K.F. and Wang, W. (2019) A review of deep learning models for time series prediction. IEEE Sensors Journal 21(6): 7833–7848. Li, S., Yang, X. and Li, R. (2019) Forecasting coal consumption in india by 2030: using linear modified linear (mgm-arima) and linear modified nonlinear (bparima) combined models. Sustainability 11(3): 695.
[15] Juan, W., Pute, W. and Xining, Z. (2013) Soil infiltration based on bp neural network and grey relational analysis. Revista Brasileira de Ciencia do Solo 37(1): 97–105.
[16] Tong, Z., Zhang, S., Yu, J., Zhang, X., Wang, B. and Zheng,W. (2023) A hybrid prediction model for catboost tomato transpiration rate based on feature extraction. Agronomy 13(9): 2371.
[17] Azam, M.F. and Younis, M.S. (2021) Multi-horizon electricity load and price forecasting using an interpretable multi-head self-attention and eemd-based framework. IEEE Access 9: 85918–85932.
[18] Ye, Z., Yin, S., Cao, Y. and Wang, Y. (2024) Ai-driven optimization of agricultural water management for enhanced sustainability. Scientific Reports 14(1): 25721.
[19] Walia, N., Singh, H. and Sharma, A. (2015) Anfis: Adaptive neuro-fuzzy inference system-a survey. International Journal of Computer Applications 123(13).
[20] McNaughton, K.G. and Jarvis, P.G. (1984) Using the penman-monteith equation predictively. Agricultural Water Management 8(1-3): 263–278.
[21] Widmoser, P. (2009) A discussion on and alternative to the penman–monteith equation. Agricultural Water Management 96(4): 711–721.
[22] Heidari, A.A., Mirjalili, S., Faris, H., Aljarah, I., Mafarja, M. and Chen, H. (2019) Harris hawks optimization: Algorithm and applications. Future Generation Computer Systems 97: 849–872.
[23] Gan, J., Lin, X., Chen, T., Fan, C., Wei, P., Li, Z., Huo, Y. et al. (2025) Improving short-term photovoltaic power generation forecasting with a bidirectional temporal convolutional network enhanced by temporal bottlenecks and attention mechanisms. Electronics 14(2): 214.
[24] Farha, Y.A. and Gall, J. (2019) Ms-tcn: Multi-stage temporal convolutional network for action segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition: 3575–3584.
[25] Li, J., Wang, X., Tu, Z. and Lyu, M.R. (2021) On the diversity of multi-head attention. Neurocomputing 454: 14–24.
[26] Shi, J., Lin, Z., Zhou, Y., Yang, Y. and Lin, J. (2025) Leveraging relation attention mechanisms for enhanced knowledge graph completion with embedding translation. EAI Endorsed Transactions on Scalable Information Systems 12(4).
[27] Yang,W., Li, S., Li, Z. and Luo, X. (2023) Highly accurate manipulator calibration via extended kalman filterincorporated residual neural network. IEEE Transactions on Industrial Informatics 19(11): 10831–10841.
[28] Wei, P., Fan, C., Yang, X., Chen, X., Gan, J., Deng, X., Wei, Y. et al. (2025) Hoes: an efficient multi-evolutionary expert system for deep learning model optimization in time series prediction. Scientific Reports .
[29] Li, L., Hu, D.Y., Tang, T.Y. and Tang, Y.L. (2022) Nondestructive testing of the quality of different grades of creamy strawberries based on laida algorithm. Journal of Food Processing and Preservation 46(11): e17008.
[30] Sun, C., Chen, Y. and Cheng, C. (2021) Imputation of missing data from offshore wind farms using spatiotemporal correlation and feature correlation. Energy 229: 120777.
[31] Hannadjas, I., James, A., Davenport, R., Lindsay, C., Brohi, K. and Cole, E. (2023) Prothrombin complex concentrate (pcc) for treatment of trauma-induced coagulopathy: systematic review and meta-analyses. Critical Care 27(1): 422.
[32] Greenacre, M., Groenen, P.J.F., Hastie, T., d’Enza, A.I., Markos, A. and Tuzhilina, E. (2022) Principal component analysis. Nature Reviews Methods Primers 2(1): 100.
[33] Schober, P., Mascha, E.J. and Vetter, T.R. (2021) Statistics from a (agreement) to z (z score): a guide to interpreting common measures of association, agreement, diagnostic accuracy, effect size, heterogeneity, and reliability in medical research. Anesthesia& Analgesia 133(6): 1633– 1641.
[34] Fan, J., Zhang, K., Huang, Y., Zhu, Y. and Chen, B. (2023) Parallel spatio-temporal attention-based tcn for multivariate time series prediction. Neural Computing and Applications 35(18): 13109–13118.
[35] Singla, P., Duhan, M. and Saroha, S. (2022) An ensemble method to forecast 24-h ahead solar irradiance using wavelet decomposition and bilstm deep learning network. Earth Science Informatics 15(1): 291–306.
[36] Bao, K., Bi, J., Gao, M., Sun, Y., Zhang, X. and Zhang,W. (2022) An improved ship trajectory prediction based on ais data using mha-bigru. Journal of Marine Science and Engineering 10(6): 804.
[37] Ghasemi, A. and Tejani, G.G. (2025) A new optimization algorithm based on the integration of reinforcement learning and hho algorithm for virtual machine placement in cloud data centers. Computing 107(6): 145.
[38] Wang, G., Tang, P., Xu, B., Zhang, C., Tian, D., Ren, J. and Li, Z. (2023) Study on water and heat transport of different types of vegetations and fields in pengbo alpine irrigation district of qinghai tibet plateau. Journal of Hydrology 618: 129201.
[39] Dai, Z.,Hu, L. and Sun, H. (2025) Robust generalized pca for enhancing discriminability and recoverability. Neural Networks 181: 106814.
[40] Al-Selwi, S.M., Hassan, M.F., Abdulkadir, S.J.,Muneer, A., Sumiea, E.H., Alqushaibi, A. and Ragab, M.G. (2024) Rnn-lstm: From applications to modeling techniques and beyond—systematic review. Journal of King Saud University - Computer and Information Sciences 36(5): 102068.
[41] Song, B., Liu, Y., Fang, J., Liu, W., Zhong, M. and Liu, X. (2024) An optimized cnn-bilstm network for bearing fault diagnosis under multiple working conditions with limited training samples. Neurocomputing 574: 127284.
[42] Deng, C., Shen, Y., Wang, Z. and Li, D. (2026) Dynamic forward–inverse prediction of excavations using mobilisable strength design and multi-receptive field convolutional neural network. Reliability Engineering & System Safety : 112454.
[43] Xie, Z., Chen, L. and Li, Y. (2026) Optimization of dam safety monitoring models based on residual compensation: A multidimensional temporal framework integrating tcn-bilstm-attention mechanism. Engineering Structures 348: 121869.
[44] Shi, Z., Xu, J., Ibrahim, A.W., He, Y., Liu, S. and Zeng, L. (2025) Accurate state of health estimation based on the multi-feature extracted and improved mha assisted cnnbigru model for lithium-ion batteries. Journal of Energy Storage 131: 117598.
[45] Sarker, P., Ksibi, A., Jamjoom, M.M., Choi, K., Nahid, A.A. and Samad, M.A. (2025) Breast cancer prediction with feature-selected xgb classifier, optimized by metaheuristic algorithms. Journal of Big Data 12(1): 78.
[46] Huang, X., Tang, J. and Shen, Y. (2024) Long time series of ocean wave prediction based on patchtst model. Ocean Engineering 301: 117572.
[47] Shen, B., Wang, Z., Ling, G., Wang, H., Cheng, X. and Miao, W. (2026) Mfma–informer: A short-term traffic flow prediction model incorporating spatiotemporal features. Journal of Transportation Engineering, Part A: Systems 152(1): 04025117.
[48] Harmer, S.L. (2025) The time machine: feedback loops, post-transcriptional regulation, and environmental integration in the plant circadian oscillator. The Plant Journal 122(6): e70275.
[49] Tadjeddine, A.A., Bendelhoum, M.S., Bendjillali, R.I., Hamiani, H. and Djelaila, S. (2023) Vre integrating in piat grid with optimal techniques: A case study kabertene. EAI Endorsed Transactions on Energy Web 10.
[50] Zhang, Q., Sheng, J., Zhang, Q., Yang, Z., Xin, Y.,Wang, B., Zhang, R. et al. (2025) Optimizing deep learning with improved harris hawks optimization for Alzheimer’s disease detection. Artificial Intelligence Review 58(10): 301.
[51] Chen, Z., Li, S., Khan, A.T. and Mirjalili, S. (2025) Competition of tribes and cooperation of members algorithm: An evolutionary computation approach for model free optimization. Expert Systems with Applications 265: 125908.
[52] Zhang, X., Zheng, F., Jiang, L. and Guo, H. (2024) Cnb net: A two-stage approach for effective image deblurring. Electronics 13(2): 404.
[53] Dai,W., Li, X., Ji,W. and He, S. (2024) Network intrusion detection method based on cnn-bilstm-attention model. IEEE Access 12: 53099–53111.
[54] Wei, P., Deng, X., Chen, X., Li, Z., Gan, J. and Shu, H. (2026) Hybrid mapping polynomial neural network based on pso optimization for robot calibration. IEEE Transactions on Instrumentation and Measurement .
[55] Liu, M., Cheng, X., Cao, Y. and Zhou, Q. (2025) Blueberry remaining shelf-life prediction based on the pso-cnn-bilstm-mha model. IEEE Access .
[56] Li, Y., Zhai, S., Yi, G., Pang, S. and Luo, X. (2025) Short term photovoltaic power forecasting based on iceemdantcn- bilstm-mha. Symmetry 17(10): 1599.
[57] Duan, C., Cao, H., Liu, F., Li, X., Duan, X. and Sheng, B. (2025) Lithium-ion batteries remaining useful life prediction using a parallel bilstm-mha neural network based on a ceemdan module. The International Journal of Advanced Manufacturing Technology 137(7): 3359–3386.
[58] Pradeepa, M. and Ponnusamy, R. (2026) An intelligent multi-head attention-driven temporal convolutional architecture for intrusion detection in wireless sensor networks. Engineering, Technology & Applied Science Research 16(1): 31490–31494.
[59] Chen, S., Li, Q., Zhao, J., Bin, Y. and Zheng, C. (2022) Neuropred-clq: incorporating deep temporal convolutional networks and multi-head attention mechanism to predict neuropeptides. Briefings in Bioinformatics 23(5): bbac319.
[60] Pang, S., Zou, L., Zhang, L., Wang, H., Wang, Y., Liu, X. and Jiang, J. (2025) A hybrid tcn-bilstm short term load forecasting model for ship electric propulsion systems combined with multi-step feature processing. Ocean Engineering 316: 119808.
[61] Yang, X., Li, B., Jia, H., Ma, Z., Sun, H., Xue, C. and Zhang, Z. (2026) A method of named entity recognition in traditional chinese medical cases by integrating bert and bilstm. Biomedical Signal Processing and Control 112: 108551.
[62] Sobuz, M.H.R., Datta, S.D., Jabin, J.A., Aditto, F.S., Hasan, N.M.S., Hasan, M., Zaman, A.A.U. et al. (2024) Assessing the influence of sugarcane bagasse ash for the production of eco-friendly concrete: Experimental and machine learning approaches. Case Studies in Construction Materials 20: e02839.
[63] Yan, J., Sun, W., Sun, B., Zhou, X., Lu, G., Li, X. and Zheng, X. (2025) Neonatal pain speech emotion recognition based on multiscale multilevel mae network. IEEE Transactions on Computational Social Systems .
[64] Song, J., Li, W., Duan, C. and Zhu, X. (2024) R2-gvio: A robust, real-time gnss-visual-inertial state estimator in urban challenging environments. IEEE Internet of Things Journal 11(12): 22269–22282.
Downloads
Published
Issue
Section
Categories
License
Copyright (c) 2025 Jinbo He, Zirun Wang, Wanzhen Huang, Fanting Zhou, Anqi Wang, Xianjin Wu, Pengzi Chen, Tongli He, Jianhong Gan, Peiyang Wei, Zhibin Li, Chunjiang Wu
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
