论文部分内容阅读
结合实测数据,以三个对数正态分布函数的和函数为拟合函数,以梯度下降法为主要方法,对沉积物粒度分布进行了数据拟合,通过数值实验我们发现:利用梯度下降法可以有效地优化分布函数的各参数,实现拟合残差的稳步持续减小,具有良好的可操作性,拟合效果是令人满意的,它为我们进行数据拟合提供了一条新的思路,同时此方法也可以推广到解决其他极值问题.
Combined with the measured data, the sum function of three logarithm normal distribution functions is taken as the fitting function, and the gradient descent method is used as the main method to fit the sediment particle size distribution. Through numerical experiments, we find that the gradient descent method Can effectively optimize the parameters of the distribution function to achieve a steady and continuous reduction of the fitting residuals, has good maneuverability, the fitting effect is satisfactory, and it provides a new idea for data fitting , At the same time this method can also be extended to solve other extreme problems.