College of Geo- Exploration Science and Technology��Jilin University��Changchun 130026��China

BP (Back Propagation) neural network and PSO (Particle Swarm Optimization) are two main heu- ristic optimization methods�� and are usually used as nonlinear inversion methods in geophysics�� The authors ap- plied BP neural network and BP neural network optimized with PSO into the inversion of 3D density interface re- spectively��and a comparison was drawn to demonstrate the inversion results�� To start with��a synthetic density interface model was created and we used the proceeding inversion methods to test their effectiveness�� And then two methods were applied into the inversion of the depth of Moho interface�� According to the results��it is clear to find that the application effect of PSO- BP is better than that of BP network�� The BP network structures used in both synthetic and field data are consistent in order to obtain preferable inversion results�� The applications in synthetic and field tests demonstrate that PSO- BP is a fast and effective method in the inversion of 3D density interface and the optimization effect is evident compared with BP neural network merely��and thus��this method has practical value��