基于正弦余弦算法的非线性数据拟合及应用  

Nonlinear Data Fitting Based on Sine Cosine Algorithm and Its Application

作　　者：雍龙泉[1,2] Yong Longquan(School of Mathematics and Computer Science,Shaanxi University of Technology,Hanzhong Shaanxi 723001,China;Shaanxi Key Laboratory of Industrial Automation,Shaanxi University of Technology,Hanzhong Shaanxi 723001,China)

机构地区：[1]陕西理工大学数学与计算机科学学院,陕西汉中723001 [2]陕西理工大学陕西省工业自动化重点实验室,陕西汉中723001

出　　处：《统计与决策》2022年第8期47-50,共4页Statistics & Decision

基　　金：国家自然科学基金资助项目(11401357);陕西省教育厅重点科学研究计划项目(20JS021);陕西省教育厅专项科研计划项目(17JK0146);陕西理工大学科研项目(SLGYQZX2002)。

摘　　要：针对非线性数据拟合问题,文章建立了以残差的平方和与绝对值和为目标的最小二乘与最小一乘模型,采用正弦余弦算法计算模型参数。结果表明,如果数据分布是对称且无异常值的,则最小二乘得到的结果与最小一乘得到的结果相差不大。最后给出一个应用实例,通过检测某人饮酒后血液中的酒精含量,得到了酒精含量随时间的变化规律。For nonlinear data fitting, this paper establishes least squares and least absolute deviation models respectively aiming for the sum of squares and sum of absolute values of residuals, and then uses sine cosine algorithm to calculate model parameters. The results show that if the data distribution is symmetric without outliers, the results obtained by least squares are not quite different from those obtained by least squares. Finally, the paper presents an application example, in which the variation pattern of alcohol content over time is obtained by measuring the blood alcohol content of a person after his drinking.

关 键 词：非线性数据拟合 最小二乘 最小一乘 正弦余弦算法 

分 类 号：O212.1[理学—概率论与数理统计]