基于 MATLAB 的土工击实试验数据处理的探讨

摘要

Because most of the existing indoor standard compaction test data processing adopts hand-painted method,the spline curve method or quadratic,cubic polynomial fitting.It is difficult to find the most value directly form the curves of Hand-painted method and spline curve method.Excellis of-ten used in quadratic and cubic polynomial fitting method,but it is a rough calculation.In order to solve above problems,the MATLAB curve fitting functions is used in this paper to analysis the com-paction test data in engineering,and based on the residuals graph of fitting curve,the maximum dry density is got accurately.The conclusions are as follows:the higher a polynomial degree of the com-paction curve is,the better fitting effect would be;and the distribution of the compaction data also will affect the result of curve fitting.%因现有的室内标准击实试验的数据处理大部分采用手绘法、样条曲线法或二次、三次多项式拟合法，手绘法和样条曲线法所绘出的曲线很难直接找到最值，对于二次、三次多项式拟合法常采用 Excel，但 Excel 计算较粗略。针对以上问题，文中运用 MATLAB 中曲线拟合的功能，对工程中击实试验数据进行分析，以拟合曲线的残差图为依据，较准确地得到了最大干密度，并得出如下结论：击实曲线多项式的次数越高，所拟合的效果越好；另外击实数据的分布情况也将影响曲线拟合效果。