Dynamic displacement is an important physical quantity in the fields of seismic engineering， military weapon design， and structural health monitoring. In the actual test process， the acceleration can usually be directly measured. Due to the uncertain test conditions such as the environment， the acceleration signal is unavoidable contains low-frequency and highfrequency noise， which causes a significant drift in velocity and displacement during the acceleration integration process. Based on the theoretical frame work of Bayesian inference， a Bayesian learning dynamic displacement identification method is constructed. The results show that， the displacement response obtained by inversion for different noise conditions （white noise， artificial noise） is basically consistent with the analytical displacement； the displacements of inversion of acceleration sensor signals with different performances are com pared by using a large shaking table test data， and their uncertainty is analyzed. The results show that this method has certain ad vantages in the characterization of the acceleration-displacement relationship， and can achieve the displacement solution without re lying on the processing of the acceleration signal， thereby avoiding the displacement integral distortion caused by the accumulated noise error.