The nonlinear dynamic differential equation of seat suspension system was built and solved with Runge-Kutta method. Taking the minimum sum of square errors of root-mean-square values of seat acceleration between simulation and test results as the objective function, the nonlinear stiffness and damping parameters of the system were identified and optimized with constraint random direct method. The identified results showed the original stiffness and damping parameters of the seat suspension were mismatched and weakened the attenuating vibration ability of the system. Using the optimized stiffness and damping parameters, the root-mean-square values of seat acceleration were brought down by 50%. To coincide with the optimized parameters, the new nonlinear rubber spring was developed with finite element method according to the ideal stiffness curve of seat suspension. So the analytical methods above have important reference value to identify and optimize the dynamic parameters of other vehicle suspension systems.