摘要
准确有效的降雨预测模型对于农业水资源管理有重要意义。文章以山西省兴县地区历史降雨资料为研究对象,提出了一种基于长短期记忆网络(LSTM)结合多特征输入的模型方法,并将预测结果与降雨实测数据进行比对验证。结果表明,引入多特征的LSTM模型相较于仅以降水量为特征的模型在评估指标上有明显提升,表现出更好的预测性能。结合多特征的LSTM模型能够满足精确预测降雨的实际需求,对于提高农业水资源管理效率具有重要意义。
Accurate and effective rainfall prediction model is of great significance for agricultural water resources management.Taking the historical rainfall data of Xingxian County in Shanxi Province as the research object,this paper proposes a model method based on long short-term memory network(LSTM)combined with multi-feature input,and compares the prediction results with the measured rainfall data.The results show that the LSTM model with multiple features has a significant improvement in the evaluation index compared with the model only characterized by precipitation,showing better prediction performance.The LSTM model combined with multiple features can meet the actual needs of accurately predicting rainfall,which is of great significance for improving the efficiency of agricultural water resources management.
作者
马良翮
MA Lianghe(Shanxi Open University,Taiyuan 030027,Shanxi,China)
出处
《工程技术研究》
2024年第6期48-50,共3页
Engineering and Technological Research
基金
山西开放大学2023年度校级课题“基于LSTM神经网络的山西地区降水量预测”(SXKDKT202312)。
关键词
降水量
农业水利
LSTM
特征输入
precipitation
agricultural water conservancy
LSTM
feature input
