非线性回归模型参数估计的区间分析方法

现有很多方法都属局部搜索方法，不能保证得到问题的全部全局最优解，而基于区间分析的区间全局优化算法则能在给定精度范围内求出问题的全部全局最优解，并能给出满足要求的包含最优解的任意小区间。基于此，给出了非线性回归模型参数估计的区间全局优化算法，论述了算法求解问题的基本思想、解算步骤、基本算法和加速工具等，并将其应用于非线性回归模型参数估计中，仿真实验结果验证了所给算法的可行性和有效性．

An Interval Analysis Algorithm for Parameter Estimation of Nonlinear Regression Model

The parameter estimation for nonlinear regression model is studied. Owing to most of the methods solved the problems are of local optimization, so they can not guarantee that the global minima have been found. However, the interval global optimization algorithm that based on interval analysis can yield guaranteed information about the global minimum and the points as well as the intervals which contain the global minimum and the points of the problem within the given precision. So the interval global optimization algorithm is present, and then the main principles, the characteristics, the basic steps and the accelerating devices of the algorithm are discussed. The application of the algorithm in the parameter estimation of nonlinear regression model by an example is analyzed. The results show the feasibility and effectiveness of the algorithm.