To address the degradation of response reconstruction accuracy and potential filter divergence caused by measurement noise covariance deviation due to uncertain weight selection in moving-window noise covariance estimation

a multi-factor noise correction Kalman filter (CMWKF) method combined with moving-window is proposed for structural response reconstruction. Initially

arbitrary empirical weights are selected

and the noise covariance is estimated using the improved moving-window method. Multiple correction factors are designed to correct the estimated measurement noise covariance. Finally

based on the Kalman filter algorithm

the responses at unmeasured locations of the structure are reconstructed using the responses from limited measurement points. The effectiveness of the proposed method is verified through case studies on a two-dimensional crane truss and a cantilever beam structure. The results show that the proposed method can effectively mitigate the influence of weight selection in moving-window noise covariance estimation and accurately reconstruct the response time-history curves at unmeasured points of the structure.