Response Reconstruction Based on Weighted Multi-innovation Kalman Filter Algorithm
The Kalman filter algorithm is easy to be affected by noise,which makes the response reconstruction accura-cy descend or even diverge.To solve this problem,a response reconstruction method based on weighted multi-innovation Kalman filter algorithm is proposed.Firstly,based on the fusion of multi-innovation theory and Kalman filter algorithm,a weighted matrix is introduced to dynamically adjust the weights of the innovation matrix to reduce the influence of historical interference accumulation.Then,the proposed method is applied to the response reconstruction,and the finite acceleration re-sponses are used to reconstruct the acceleration,velocity,and displacement responses at the remaining unknown positions.Finally,numerical simulation and experimental verification are carried out on a crane truss and a simply supported beam re-spectively.The results show that compared with the response reconstruction methods based on the Kalman filter algorithm and the multi-innovation Kalman filter algorithm,the proposed method has better filtering stability and estimation accuracy,and can achieve higher reconstruction accuracy with little increase in operation time.
vibration and wavecraneresponse reconstructionKalman filtering algorithmmulti-innovation theoryweighted multiple innovation
万方数据
维普
