地球物理资料非线性反演方法讲座(九)——蚁群算法  

Lecture on Non-Linear Inverse Methods in Geophysical Data(9)——Ant Colony Optimization

作　　者：王书明[1] 刘玉兰 王家映[3] 

机构地区：[1]中国地质大学地球物理与空间信息学院,武汉430074 [2]东方地球物理勘探有限责任公司,涿州072751 [3]中国地质大学 地球物理与空间信息学院,武汉430074

出　　处：《工程地球物理学报》2009年第2期131-136,共6页Chinese Journal of Engineering Geophysics

摘　　要：蚁群算法是一种仿生类非线性优化算法,具有并行性、正反馈性和全局极小搜索能力强等特点。蚁群算法的机理是:生物界中的蚂蚁在搜寻食物源时,能在其走过的路径上释放一种蚂蚁特有的分泌物-信息素,使得一定范围内的其他蚂蚁能够觉察并影响其行为。当某些路径上走过的蚂蚁越来越多时,留下的这种信息素轨迹也越多,以至信息素强度增大,使后来蚂蚁选择该路径的概率也越高,从而更增加了该路径的信息素强度。为了将起源于离散网络路径优化的原始蚁群算法思想用于连续函数优化的地球物理反演问题,必须对有关实施细节进行改造和修正,本文基于网格划分策略的连续域蚁群算法实现了连续域大地电磁蚁群算法。通过选择蚂蚁数、信息素挥发系数等参数,利用三层K型模型和四层HA型模型进行数值试验,结果表明,蚁群算法可以稳定收敛,反演结果接近理论模型。Ant colony optimization is a kind of bionics nonlinear optimization algorithm. The principle of ant colony optimization is： when one ant finds a good path from the colony to a food source, a kind of pheromone trail starts to evaporate, and then the other ants are more likely to follow that path. The more time it takes for an ant to travel down the path and back again, the more time the pheromones have to evaporate. A short path, by comparison, gets marched over faster, and thus the pheromone density remains high as it is laid on the path as fast as it can evaporate. Based on continuous domain ant colony algorithm with network division strategy, magnetotelluric ant colony inversion in continuous domain has been carried out in this paper. By taking characteristics of geophysics inversion into account, this paper successfully actualized magnetotelluric ant colony inversion in continuous domain. By choosing suitable inversion parameters such as a number of ants and pheromone volatilization coefficient, the numerical tests were done through taking use of 3--layer K models and 4--layer HA models. The results show that ant colony optimization can be stably converged, and the results of inversion are close to the theoretical models.

关 键 词：蚁群算法 非线性 反演 地球物理 

分 类 号：P631[天文地球—地质矿产勘探]