遗传算法在多向随机波浪反射分析中的应用

通过数值模拟方法研究了遗传算法在分析多向入、反射波浪共存场中入射波浪的方向分布、波浪的反射系数和反射面位置的应用.对入射波浪的方向分布、波浪的反射系数和反射面位置通过遗传优化模拟和理论互谱的差别组成适应函数来确定.数值模拟结果表明,在多向入、反射波浪场中遗传算法能够精确地分离入、反射波浪,而且不需要预先确定反射面位置,因此遗传算法能够很好地被应用于分析多向随机波浪从斜坡堤等反射面位置不明确的建筑物的反射问题.

The application of genetic algorithm for multi-directional random wave reflection analysis

The validity of the genetic algorithm in estimating the directional spreading, reflection coefficient and reflective line distance is studied by numerical simulation. The genetic algorithm is verified by the numerical simulation of wave cross-spectrum and surface elevations. The directional spreading of the incident wave, reflection coefficient and reflection line distance are obtained by optimizing a fitness parameter which is a function of the difference between the simulated and theoretical cross-spectra. The numerical simulation results show that the genetic algorithm can accurately separate the incident and reflected wave in multi-directional reflective wave fields without a prior knowledge of the reflection line position. The analytic and model test results indicate that the genetic algorithm can be applied to analyzing the multi-directional random wave reflection.