AVO岩性参数反演的改进人工鱼群算法研究

为及时获取地震岩性参数,提出了基于动物自治体模型的人工鱼群算法进行参数反演。该方法对鱼群算法采取分阶段策略进行改进,并增加了跳跃与吞食行为,从而使鱼群更容易跳出局部最优得到性能优化。对振幅随炮检距变化(AVO)的实际数据参数反演的结果表明,与标准人工鱼群算法相比,改进的鱼群算法的反演精度与寻优时间都得到很大改进,表现出更强的寻优泛化能力。

Research on the Improved Artificial Fish Swarm Algorithm AVO Lithological Parameter Inversion

To get the timely seismic lithology parameters, the improved artificial fish algorithm based on animal commune model was put forward for parameter inversion. The method adopts strategies in stages to improve the fish algorithm, and increases the jump and eating behavior, thus make it easier to jump out of local optimal value for the performance optimization. Combined with the multi-parameter amplitude changing with offset (AVO) for parameter inversion of real data, the results show that compared with the standard artificial fish algorithm, the improved fish algorithm improves the precision of inversion, shortens the time of optimization, enhances the global search ability, and showes a strong optimization ability.