【摘要】 采用遗传程序设计方法来求解偏微分方程的参数识别问题,用遗传程序设计优化参数的模型结构,成功实现了对偏微分方程参数的反演。数值实验表明,采用此算法能使计算机自动搜索到满足条件的解,该方法的研究可为偏微分方程的参数识别提供一个有效的工具。更多还原

【Abstract】 This paper proposes a new way of solving parameter identification by employing genetic programming(GP). GP is employed to optimize the structure of a model and to search for the best approximate solution at the same time. It has taken the first step in successfully automating the modeling process of parameter identification for partial differential equations inverse problems. GP can optimize undetermined parameters very well as well as give good predictions, and its operations are much simpler than traditional mathematical methods. It is shown that GP has great potential to provide a new and powerful tool for parameter identification of partial differential equations.更多还原