欢迎访问《中国农学通报》，
中国农学通报

中国农学通报 ›› 2011, Vol. 27 ›› Issue (8): 280-283.

• 资源 环境 生态 土壤 气象 • 上一篇    下一篇

基于人工神经网络理论的土壤水分预测研究

宰松梅 郭冬冬 韩启彪 温季   

  • 收稿日期:2010-09-08 修回日期:2010-10-07 出版日期:2011-04-20 发布日期:2011-04-20
  • 基金资助:

    地下滴灌关键技术研究与设备开发;微压灌水器和过滤装置的中试与转化

Soil Moisture Prediction Based on Artificial Neural Network Model

  • Received:2010-09-08 Revised:2010-10-07 Online:2011-04-20 Published:2011-04-20

PDF (PC)

999

摘要：

土壤水分含量是影响作物生长的重要因素，精确的预测技术对水资源的合理利用与管理具有重要的指导意义。利用人工神经网络理论，建立了以降水量、蒸发量、相对湿度和地下水埋深为输入因子，土壤水分含量为输出因子的预测模型，并对其预测精度进行了评价。结果表明，BP神经网络模型预测土壤含水率的最大误差为8.66%，平均误差为4.27%，预测精度达到0.989。模型具有较高的预测精度，其结果可为制定合理的水资源调配方案和调度计划提供科学依据。

关键词: 茶园土壤, 茶园土壤, 三叶草, 稻草覆盖, 杂草生物多样性

Abstract:

Soil water content is an important factor affecting crop growth, and the accurate prediction of water resources is an important guiding on their reasonable utilization and management. An artificial neural network model was established, with rainfall, evaporation, relative humidity and groundwater table as the input factors, and soil moisture as the output factors and its prediction accuracy was evaluated in this paper. The results showed that the maximum error of predicting soil moisture for BP neural network model was 8.66%, average error was 4.27%, and prediction accuracy of 0.989. BP neural network model had higher prediction accuracy for the prediction of soil moisture. The results can be used for the allocation of irrigation water resources.

中图分类号: