油茶树树龄数学模型研究
Research on the mathematical model of oil‑tea tree age
-
摘要: 树龄对油茶籽产量和茶油品质存在显著影响,为实现无损测定老油茶树树龄,建立了老油茶树树龄与地径之间的数学模型。首先,测量每株被采伐油茶树的树龄和地径;然后,以2/3的数据作为建模数据集,分别建立树龄与地径之间的线性模型、对数模型、指数模型和混合模型;最后,以剩下1/3的数据作为测试数据集,测试这4个数学模型的预测精度。线性模型、对数模型、指数模型和混合模型的校正决定系数分别为0.756、0.769、0.661和0.787,平均预测精度分别为83.379%、83.441%、78.207%和84.878%,预估标准差分别为5.441 a、6.826 a、6.843 a和5.898 a。结果表明,在这4个模型中,混合模型拟合效果最好,预估精度最高。Abstract: Tree age influences on the yield of oil-tea seeds and the quality of tea oil, to nondestructively measure the age of old oil-tea tree, a mathematical model between the age of old oil-tea tree and the ground diameter of the ground was established. First, the age and ground diameter of each harvested oil-tea tree were measured, and then 2/3 of the data were used as the modeling data set to establish the linear, logarithmic, exponential, and hybrid models between age and ground diameter, respectively, and finally, the remaining 1/3 of the data were used as the testing data set to test the prediction accuracy for these four mathematical models. The adjusted coefficients of determination for the linear, logarithmic, exponential and mixed models were 0.756, 0.769, 0.661 and 0.787, respectively, and the average prediction accuracies were 83.379%, 83.441%, 78.207% and 84.878%, with the prediction standard deviations being 5.441 a, 6.826 a, 6.843 a and 5.898 a, respectively. The results showed that, among these four models, the hybrid model exhibited the best fit and the highest prediction accuracy.