中文摘要：

将最小二乘支持向量机方法,应用于一维对流扩散方程逆过程研究。相对于其它已有的方法,该方法具有简便实用、稳定性好等优点。使用高斯核函数对方程的解进行整体逼近,利用样本点得到回归参数的值。数值结果表明,该方法具有较高的精度和稳定性,可以有效地求解其它反问题。

英文摘要：

Least square support vector machines method is used to study the inverse process of one-di- mensional convection-diffusion equation. The method has many advantages compared with the other meth- ods, such as simple, practical and good stability. The solution of one-dimensional convection-diffusion e- quation is globally approximated by use of Gauss kernel function. The values of regression parameters are estimated through sample points. This method has been successfully tested on practical examples and has yielded higher accuracy and stability solutions, so it can be used to other inverse problems.