基于SRA-GP的水库年末水位预测模型

Prediction Model of the Level at the End of Year on SRA-GP

为了克服传统预测方法的不足,利用逐步回归分析优选因子的功能和遗传程序设计搜索模型结构能力强的优点,建立了多年调节水库年末消落水位预测模型,并将其用于多年调节水库年末消落水位的预测。模型通过逐步回归分析寻找因子,再利用演化计算自动寻找最优的模型结构,与传统方法相比不仅具有较大的灵活性和智能性,而且在一定程度上提高了遗传程序的寻优效率。实例表明,模型适应性强,预测精度高,可以为水库调度提供一定依据。

In order to overcome the disadvantages of traditional method, the paper makes use of the function of stepwise regression analysis and the merit of genetic programming, establishes the hybrid SRA-GP for predicting the year end level of multi-year regulating storage reservoir. The model that uses SRA to search for factors and GP to seek for structure of model, not only is more flexible and intelligent than traditional methods can find the optimal structure automatically, but also improve the efficiency. The example of Hong-Jia-Du station on the Wu River is used to illustrate that the method has high accuracy and is effective. It could provide reference for reservoir dispatching.