位场反演是地球物理反演问题研究的主要内容之一,目前,局部和全局最优化技术结合是地球物理反演的热点。人工鱼群算法是源于对鱼群觅食行为而提出的一种新型群体智能优化算法,在多参数、非线性、多极值函数优化问题中具有较强的优越性。通过分析人工鱼群算法的原理,本文用该非线性算法实现了重磁位场反演,并且在理论模型的实验中,证明了算法的抗噪能力。 Field inversion is one of the main contents of geophysical inversion research. At present, the combination of local and global optimization techniques is a hotspot in geophysical inversion. Artificial fish swarm algorithm is a new swarm intelligence optimization algorithm which is derived from the foraging behavior of fish swarm. It has strong superiority in multi-parameter, nonlinear and multi-extremal function optimization problems. By analyzing the principle of artificial fish swarm algorithm, this paper realizes the gravity and magnetic field inversion with this nonlinear algorithm. And in the experiment of theoretical model, the anti-noise ability of the algorithm is proved.