论文部分内容阅读
大豆种子均径是育种专用排种盘型孔设计的重要依据,精准构建大豆种子均径预测模型具有非常重要的意义。利用线性回归模型、粒重物理模型及二次回归模型分别对大豆种子均径进行预测,并对其预测误差进行了对比分析。在此基础上,应用Shapley值法确定组合预测模型中各单一预测模型的权重,依此构建了大豆种子均径的组合预测模型,并对随机选取的5粒东农52大豆样本均径进行了预测。结果表明:组合预测模型预测精度高于选定的各单一预测模型,平均预测误差较小,且预测误差波动有所降低,对大豆种子均径的预测是可行且有效的。同时,为其他作物种子的相关物理参数预测提供了一种实用的新方法。
The average diameter of soybean seed is an important basis for the design of the seed-tray hole for breeding seed. It is of great importance to construct the prediction model of soybean seed average diameter accurately. The linear regression model, the gravimetric model and the quadratic regression model were respectively used to predict the average seed size of soybean, and the prediction errors were compared. On this basis, Shapley value method is used to determine the weight of each single prediction model in the combined forecasting model. Based on this, a combined forecasting model of soybean seed mean diameter is constructed, and five samples of Dongnong 52 are randomly selected prediction. The results show that the prediction accuracy of the combined forecasting model is higher than that of the single forecasting model. The average forecasting error is small and the forecasting error fluctuates lower. The prediction of the average seed size of soybean is feasible and effective. At the same time, it provides a practical and new method for the prediction of the related physical parameters of other crop seeds.