吉林农业大学学报2025，Vol.47Issue(1)：179-184,6.DOI:10.13327/j.jjlau.2024.1081
基于BAS-BP网络的土壤湿度预测方法研究
Soil Moisture Prediction Method Based on BAS-BP Network
摘要
Abstract
Meteorological factors and time series are mainly utilized in the current short-term pre-diction of soil moisture,and there are problems such as high input dimensions and insufficient preci-sion of the prediction model.Aiming at the above problems and the defects of BP neural network,this paper proposed a prediction model based on BAS-BP neural network.Weather forecasts were simu-lated using actual meteorological data to predict soil moisture and the predictions were validated and tested with the actual measured 40 cm vertical average soil moisture measured in Shuangyang Dis-trict,Changchun City.The results show show BAS-BP neural network has higher prediction accu-racy and faster convergence speed than traditional BP neural network.At the same time,compared with traditional GA-BP and PSO-BP models,it is found that BAS-BP has better performance.The easily accessible weather forecast data can be combined to accurately predict the changes in soil moisture in the next five days,which can provide scientific guidance for the utilization of agricultural water resources.关键词
土壤湿度预测/BAS算法/BP神经网络/气象因子Key words
soil moisture prediction/BAS algorithm/BP neural network/meteorological factor分类
农业科技引用本文复制引用
孟楚,李士军,常晶,穆叶,肖培,张鑫..基于BAS-BP网络的土壤湿度预测方法研究[J].吉林农业大学学报,2025,47(1):179-184,6.基金项目
吉林省教育厅项目(JJKH20220322KJ),吉林省科技发展计划项目(20210302009NC) （JJKH20220322KJ）
