瑞雷面波频散曲线的粒子群蚁群混合优化反演

应用瑞雷面波频散曲线反演地下介质的横波速度剖面是面波勘探的重要步骤之一。传统线性反演方法已不能满足物探工程的要求,非线性的反演方法成为研究热点。文中将基于粒子群优化算法和蚁群优化算法的非线性混合优化算法应用于瑞雷面波频散曲线反演,获得横波速度剖面。该算法利用信息素引导机制更新粒子的早期位置,充分结合了粒子群优化算法对全局最优解的引导策略和蚁群优化算法的局部搜索能力,克服了粒子群算法在群体处于平衡状态时粒子群更新停滞不前和蚁群算法对多极值函数求解时收敛早熟的缺点。通过对多种理论模型频散曲线的反演,检验了该算法的有效性和稳定性;与单独的蚁群算法、粒子群算法反演结果的对比验证了该算法的优越性;实测数据反演结果检验了算法的实用性。

Inversion of Rayleigh wave dispersion curves based on particle swarm and ant colony hybrid optimization

Inversion of the Rayleigh surface-wave dispersion curve to obtain the shear-wave (S-wave) velocity profile of an underground medium is one of the important steps in surface-wave exploration. The traditional linear inversion method cannot meet the needs of geophysical exploration, and the nonlinear inversion method has instead become a research hotspot. In this paper, a nonlinear optimization algorithm based on the particle swarm optimization (PSO) algorithm and the ant colony optimization (ACO) algorithm is applied to the nonlinear inversion of the Rayleigh surface-wave dispersion curve to obtain the underground S-wave velocity profile. This algorithm uses the pheromone guidance mechanism to update the positions of particles in the early stage. It fully combines the guidance strategy of the PSO algorithm for the global optimal solution with the local search ability of the ACO algorithm. Meanwhile, it overcomes the shortcoming of the PSO algorithm that particle swarm update comes to a standstill when the population is in a state of equilibrium and the defect of the ACO algorithm that convergence is premature when it is applied to solve the multi-extremum function. The effectiveness and stability of the proposed algorithm are examined by the inversion of the dispersion curves of various theoretical mo-dels; the comparison of the inversion results of this algorithm with those of the ACO and PSO algorithms alone verifies its superiority; Inversion results of measured data further test the practicability of this algorithm.