With application of particle swarm optimization to lunar geophysical parameters inversion, problems resulted from limitation of lunar seismic network and/or unsuitability of conventional inversion methods are avoided. Parameters investigated include load ratio, crustal thickness, crustal density, subsurface load depth and lithospheric elastic thickness. It indicates that the approach is able to calculate model parameters best fitting observed ones automatically and simultaneously. It performs better in computational speed. It is therefore quite useful in lunar multi-parameters inversion and can be employed in lunar interior precision structure research.