| 引用本文: | 邓轩盈,吕辛未,郑文燕,等.参考作物腾发量预报在线训练深度学习模型[J].灌溉排水学报,2024,43(12):57-64. |
| DENG Xuanying,LYU Xinwei,ZHENG Wenyan,et al.参考作物腾发量预报在线训练深度学习模型[J].灌溉排水学报,2024,43(12):57-64. |
|
| 摘要: |
| 【目的】探究参考作物腾发量（ET0）的实时预报方法。【方法】以浙江省杭州市萧山区2021年4月24日—2023年12月31日的天气预报数据和整点天气实况资料为数据集，分析模型输入数据的预报精度，采用BP神经网络算法构建ET0预报的深度学习模型，并部署至阿里云服务器进行在线训练。【结果】模型的输入数据中，气温预报准确率较高，且最低气温预报精度高于最高气温，天气类型及风力等级预报存在一定误差。模型预报值与实时数据计算得到的标准值相比，预见期内二者变化趋势大致相同，预报精度较高，训练期与测试期准确率最高分别可达到91.56%和84.75%，训练期均方根误差（RMSE）与平均绝对误差（MAE）平均值分别为0.828 mm/d和0.667 mm/d，测试期RMSE与MAE平均值分别为1.049 mm/d和0.829 mm/d。【结论】采用公共天气预报数据构建BP模型在线训练，能够实现ET0的实时预报，精度较高且便于运用，可为农业工作者实时灌溉决策提供数据支撑。 |
| 关键词: 参考作物腾发量；BP神经网络；公共天气预报；ET0预报；在线训练 |
| DOI：10.13522/j.cnki.ggps.2024176 |
| 分类号: |
| 基金项目: |
|
| Forecasting reference crop evapotranspiration using deep learning model and online training |
|
DENG Xuanying, LYU Xinwei, ZHENG Wenyan, ZHENG Shizong, ZHANG Yadong, LUO Tongyuan, CUI Yuanlai, LUO Yufeng
|
|
1. State Key Laboratory of Water Resources Engineering and Management, Wuhan University, Wuhan 430072, China; 2. Baidu Online Network Technology (Beijing) Co., Ltd, Beijing 100085, China; 3. Gansu Provincial Institute of Water Conservancy, Lanzhou 730000, China; 4. Zhejiang Institute of Hydraulics and Estuary (Zhejiang Institute of Marine Planning and Design), Hangzhou 310020, China
|
| Abstract: |
| 【Objective】Reference crop evapotranspiration (ET0) is a critical parameter for irrigation and water management. This paper proposes a method for real-time forecasting ET0 using weather forecast data and a deep learning approach.【Method】The study was conducted in Xiaoshan District, Hangzhou City, Zhejiang Province. Hourly measured weather data and 1-7 day forecasted weather data from April 24, 2021 to December 31, 2023 were used as the dataset. The forecasting accuracy of the weather data was analyzed. A deep learning model based on the backpropagation (BP) neural network algorithm was developed and deployed for online training using Alibaba Cloud servers.【Result】The accuracy of the input parameters was generally reliable, with minimum temperature forecasts being more accurate than maximum temperature forecasts. Forecasting accuracy decreased as the lead time increased. Errors were observed in forecasting weather types and wind scales. The ET0 predicted by the model closely matched those calculated using real-time data, demonstrating high forecasting accuracy. During the training period, the model achieved a maximum accuracy of 91.56%, with an average root mean square error (RMSE) of 0.828 mm/day and a mean absolute error (MAE) of 0.667 mm/day. During the testing period, the model achieved an accuracy of 84.75%, with the average RMSE and MAE being 1.049 mm/day and 0.829 mm/day, respectively.【Conclusion】By using publicly accessible weather forecast data and an online-trained BP neural network model, real-time ET0 forecast can be achieved with high accuracy. This approach offers valuable support for farmers, enabling informed and timely irrigation decisions. |
| Key words: reference crop evapotranspiration; BP neural network; public weather forecast; real-time ET0 forecasting; online training |
