基于遗传算法推求城市暴雨强度公式的工程实践

Engineering Practice of Calculating Formula of Urban Rainstorm Intensity Based on Genetic Algorithm

摘要:

以A市过去30 a的降雨资料为基础，运用优化适线法对理论频率分布曲线进行参数估算，并借助遗传算法求解超定非线性方程，分别构建了A市的单次暴雨重现期公式和暴雨总量公式。结果表明：相较于传统手段，利用遗传算法推求暴雨公式可大幅缩减人力和时间投入，通过均方根误差和相对均方根误差来评判这些公式的精度也完全符合相关规范的要求，可为各地暴雨公式修订工作提供一定的借鉴。

Abstract:

Based on the rainfall data of City A over the past 30 a, the parameters of the theoretical frequency distribution curve are estimated by using the optimized curve-fitting method, and the over-determined nonlinear equation is solved with the aid of the genetic algorithm. The formulas for the recurrence period of single rainstorm and the total amount of rainstorm in City A are constructed respectively. The results indicate that compared to traditional methods, the use of the genetic algorithm to derive the rainstorm formulas significantly reduces the human and time investments. The accuracy of these formulas is judged through the root mean square error and relative root mean square error, which also fully complies with the requirements of relevant norms, and can provide a certain reference for revising the rainstorm formulas in various regions.