System matrices identification with Riccati equation solution and global optimization to reduce experimental uncertainties

In this paper, an inverse system identification method based on actual structural vibration response data is proposed. The mass, stiffness and damping matrices of the equivalent system are established based on the properness condition of the modes. Aiming at the modal properness conditions of eigenvector, the eigenvector is first modified based on the solution of the Riccati equation in control theory. However, the resulting frequency response function is different from the actual frequency response function. In this case, the objective function is defined by the difference between the initial frequency response function and the simulated frequency response function obtained after the system matrices reconstruction. New optimization methods such as genetic algorithm are used to optimize the frequency response function to obtain a more accurate frequency response function. This paper presents a simulation case to illustrate the above methods. The optimization results show that the above methods can provide more accurate modal parameters.