摘要
通过6个典型测试函数对一种新型蝙蝠算法(NBA)进行仿真验证,并与基本蝙蝠算法(BA)的寻优结果进行对比。利用NBA算法对承压含水层参数进行反演,并以2个实例进行分析研究。结果表明:NBA算法寻优精度优于BA算法15个数量级以上,具有较好的收敛精度和全局寻优能力,能有效避免基本BA算法陷入局部最优;NBA算法对2个实例的反演精度较文献改进SA算法等提高了56.5%以上,获得了较好的反演效果。利用该算法对承压含水层参数进行反演,不但为精确估计承压含水层参数提供了有效方法,而且拓展了NBA算法在地下水模型参数反演中的应用,具有良好的前景。
Through 6 typical test functions, a new type of bat algorithm( NBA) is simulated and verified, and the results are compared with the results of basic bat algorithm( BA). NBA algorithm is used to inverse the parameters of confined aquifer, and the analysis is carried out with 2 examples. The results show that: 1NBA algorithm has better convergence accuracy and global optimization ability,and can effectively avoid the local optimum of the basic BA algorithm. 2NBA algorithm for 2 examples of the inversion accuracy of the improved SA algorithm and other improved more than 56. 5%, obtained a good inversion results. Using this algorithm, the parameters of confined aquifer can not only provide an effective method for accurate estimation of confined aquifer parameters, but also expand the application of NBA algorithm in parameter inversion of groundwater model.
出处
《人民珠江》
2016年第12期75-81,共7页
Pearl River
