将三峡过闸船舶的吃水和载货量视作一组离散的数据,对船舶最大吃水、定额吨和实载吨等3个变量之间的关系进行拟合分析,提出运用插值方法确定实际过闸船舶的最优吃水。对连续观测的三峡船闸2014年的22065艘次下行过闸船舶的数据进行预处理,构建船闸门槛水深与允许过闸船舶最大吃水的计算模型。运用数理统计对8个区间的变量值进行聚类分布。运用MATLAB中分段线性、边界条件为Lagrange和二阶导数的三次样条插值运算方法对样本数据进行拟合。结果表明:以二阶导数为边界条件的三次样条插值可解决该问题,当允许过闸船舶的吃水控制在3.92~4.38m时,过闸船舶装载率为最佳。
Fitting analysis is performed with the discrete data set of vessel maximum draft, rated tonnage and real load tonnage from actual statistics of the Three Gorges Lock. An interpolation method is brought up to determine the optimized draft. For the processing, the model which relates the vessel maximum draft to the water depth on ship-lock sill is built on the basis of 22 065 down-going vessel data acquired in 2014. The variables are clustered as eight sections. A piecewise linear three-order interpolation is carried out to find the relationship among variables with the Lagrange and second derivatives related boundary conditions. The fitted curve by cubic spline interpolation with the boundary conditions of second derivatives is shown valid. It concludes that vessels at drafts among 3.92 and 4.38 meters will have the optimal loading rate.