飞蛾火焰优化算法在马斯京根模型参数优化中的应用

Application of Moth-flame Optimization Algorithmic to Optimization of Parameter in Muskingum Model

通过10个典型二维函数对一种新型群体智能仿生算法——飞蛾火焰优化(MFO)算法进行仿真验证,并与粒子群优化(PSO)算法的寻优结果进行对比。利用该算法优化马斯京根模型参数,并以相关文献中的3个实例进行验证。结果表明:MFO算法在二维函数极值寻优问题上具有较好的收敛精度和全局寻优能力,寻优精度较PSO算法提高了7个数量级以上。利用MFO算法优化马斯京根模型参数,可以获得比相关文献更高的模拟精度,为精确估计马斯京根模型参数提供了有效方法。

In this paper, 10 typical two-dimensional functions are used to validate a new kind of swarm intelligent bionic algorithm-moths-flame optimization( MFO) algorithm. The result is compared with the optimization results that of and the particle swarm optimization( PSO). This algorithm is used to obtain the optimal parameter of Muskingum model algorithm, and verified by 3 examples in the relevant literature. The results show that the MFO algorithm has better convergence precision and global optimization ability in the optimization problem of the two dimensional function extremum, and the optimization accuracy is improved by 7 orders of magnitude compared with the PSO algorithm. The simulation accuracy of MFO algorithm Muskingum model parameter optimization is higher than the relevant literature, and can provide the effective method for the accurate estimation of Muskingum model parameters.