基于改进遗传规划算法的数据拟合*

针对传统数据拟合方法需预先估计基函数、依赖于应用领域等问题，基于遗传规划的动态可变特性，提出将遗传规划与最小二乘法结合，设计具有一定通用性和自适应能力的数据拟合算法。在分析传统遗传规划算法的基础上，详细介绍了算法改进方法，并针对各种类型的拟合数据进行了对比实验。实验结果表明，该算法不仅可以应用到多种场合，而且可以提高拟合效率与精度。

Data fitting based on improved genetic programming

There are many problems in current data fitting methods, such as it needs to estimate the radical function in advance and depends on the application field, and so on. Based on the dynamic alterable property of genetic programming (GP),combined GP with least square method, and designed a new data fitting method which had universal and self-adaptive capacity. Firstly,analyzed the traditional GP.Secondly,introduced the improved method in details.Finally,fimiched some contrastive experiments based on various fitting data. Experiment results show that this method can be applied in many fields, and it can improve the fitting efficiency and precision.