基于Matlab的线性拟合计算  

Study on Linear Fitting Algorithm Based on Matlab

作　　者：田春军[1] 刘丕德[1] 

机构地区：[1]中船重工集团公司江苏自动化研究所,江苏连云港222006

出　　处：《情报指挥控制系统与仿真技术》2005年第6期14-17,共4页Information Command Control System and Simulation Technology

摘　　要：对于弹道导弹,由于其运动受多种因素的影响,因此描述其运动的微分方程是极其复杂的,而且,方程中的部分参数是通过对试验数据的科学处理获得的。为了建立数学模型,有时用数据拟合的方法对试验数据进行处理。运用单变量、多变量的线性拟和方法,给出了“单输入/单输出”、“多输入/单输出”的线性拟合模型,并针对拟合模型中未知参数的确定问题,给出了基于最小二乘的线性矛盾方程组的求解方法。通过仿真计算,能够得到较高精度的计算结果。最后,给出了Matlab的实现方法,并给出了一段具体的实现程序。Trajectory differential equation that describes ballistic missile movement is influenced by many factors, Some parameters of trajectory movement is very complicated as its differential equation are obtained by testing data processing. For establishing mathematical model, the method of data fitting is adopted to process testing data. The linear fitting model of “single-input/single-output” ＆ “multi-input/single-output” is given by using the univariable ＆ multivariable fitting method. In view of the unknown parameters of the fitting model, it gives the algorithm of linear contradiction functions based on least-square. The simulation demonstrates that the resolving precision is very high. Finally, the implementation method based on Matlab and the implementation program are given.

关 键 词：MATLAB 线性拟合 单变量拟合 多变量拟合 计算 

分 类 号：O241.6[理学—计算数学]