摘要
基于遗传程序拟合非线性函数的思想,建立了量化计算作物干旱程度的遗传程序模型。该方法通过演化计算自动寻找最优的模型结构,不但可以准确地反映出干旱对作物造成的产量损失,而且避免了预先建立具体数学表达式及求解参数的不便。实例表明:遗传程序设计的作物干旱程度量化模型,具有较大的灵活性和智能性,拟合精度高,可为农业干旱风险评估提供科学的依据和坚实的基础。
Based on the thought of using genetic programming for nonlinear function fitting, a genetic programming model, which can quantificationally reflect the crop drought degree, is established in this paper. The model can automatically find the optimal model structure through evolutionary algorithm. So it not only can accurately reflect the crop yield loss caused by drought, but also can avoid the inconvenience of establishing concrete expression and solving parameters. Research shows that the crop drought degree evaluation model based on genetic programming has high accuracy, flexibility and intelligence. The model can supply scientific basis for the risk evaluation of agricultural drought.
出处
《节水灌溉》
北大核心
2008年第1期41-43,共3页
Water Saving Irrigation
基金
河南省高校创新人才培养工程项目(HNCX2003-17)