基于LSTM神经网络的降水量预测研究
Research on Precipitation Prediction Based on LSTM Neural Network
马良翮1
作者信息
- 1. 山西开放大学,山西 太原 030027
- 折叠
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.
Abstract
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.
关键词
降水量/农业水利/LSTM/特征输入Key words
precipitation/agricultural water conservancy/LSTM/feature input引用本文复制引用
基金项目
山西开放大学校级项目(2023)(SXKDKT202312)
出版年
2024
NETL
NSTL
万方数据