土壤温度时间序列预测的BP神经网络模型研究  

Application of BP neural network in forecasting soil temperature time-series

作　　者：邹平[1] 杨劲松[1] 姚荣江[1] 

机构地区：[1]中国科学院南京土壤研究所,南京210008

出　　处：《中国生态农业学报》2008年第4期835-838,共4页Chinese Journal of Eco-Agriculture

基　　金：中国科学院知识创新工程重要方向项目(KZCX2-YW-406-3);国家高技术研究发展计划(863计划)课题(2007AA091702;2006AA10A301);国家科技支撑计划课题(2006BAD05B04;2006BAD05B02);国家自然科学基金项目(40771097)资助

摘　　要：针对滨海盐渍区表层土壤温度时序变化复杂、高度非线性的特点,以江苏省苏北典型滩涂区域为研究对象,综合运用BP神经网络和时间序列多维拓展的方法,对长期定位监测点表土层土壤温度时间序列数据进行了分析和预测,为土壤溶质运移研究与当地作物合理布局提供理论基础和参考依据。结果表明,输入层、隐含层和输出层神经元数目分别为7、7和1的3层BP神经网络模型用于土壤温度时间序列训练仿真时效果最优,其误差平方和达最小值18.017。选定的此结构BP神经网络模型简单、实用,有良好的推广泛化能力,经独立测试样本检验,预测值与实测值的相对误差均在20%以内,平均相对误差仅为2.94%,可满足土壤温度日常预报的需要。Soil temperature is an important parameter in agro-meteorological observations and research on variations in soil temperature time-series provides the theoretical basis for soil solute movement and proper distribution of reference crops. Owing to the complicated characteristics and strong nonlinearity of surface soil temperature time-series in saline coastal regions, a long-term da- ta on soil temperature was analyzed and soil temperature time-series predicted by integrated Back-Propagatlon （BP） neural network method and expansion of multi-dimensional time-series, using experimental data from a typical coastal region in North Jiangsu Prov- ince. The results show that BP neural network model with 7 neurons of input layer, 7 neurons of hidden layer, and 1 neuron of output layer is the best for soil temperature time-series forecast, with a minimum sum squared error of 18. 017. Relative errors between forecated value and measured value of the soil temperature all fall within 0 - 20% , with an average relative error of only 2.94%. The high range of relative error underscores the significance of BP neural network in daily soil temperature forecasting.

关 键 词：土壤温度预测 时间序列 BP神经网络 滨海盐碱区 农业气象观测 

分 类 号：S152.8[农业科学—土壤学]