Research On Accuracy And Sensitivity Of Dynamic Load Identification In Time Domain

With the development of dynamic load identification technology,more and more dynamic load identification methods are applied to practical engineering.However,how to improve the accuracy and scope of load identification has become one of the important issues that people care about.This paper focuses on the Wilson-? inverse analysis method,from the simple to complex principle,from the single degree of freedom system,the multi-degree of freedom system to the continuous beam system,studies the sensitivity of the dynamic load identification method to various parameters,and puts forward the practical a viable correction method.In this paper,the influence of frequency,phase and amplitude of the interference on the identification accuracy of the dynamic load is analyzed with the fixed-frequency interference as the breakthrough.It is found that the interference amplitude has a great influence on the recognition result,the influence of the interference phase is small,and what most worth exploring is that the influence of the interference frequency on the load identification.In addition,this paper proposes an improved numerical correction algorithm for variable-interval time-domain dynamic load identification,which can well overcome the problem that the original algorithm is unstable under various frequencies due to improper selection of iterative interval.The method has high precision.The simulation and experimental results show that the method is stable and accurate,and has certain practical application value.This paper also studies the influence of model error on the accuracy of load identification.It is found that the error of each parameter has different effects on the accuracy of load recognition,but it follows a certain law.The influence of model local parameter error on the load identification result is generally smaller than the overall parameter error of the model.In addition,this paper proposes a model correction algorithm based on Wilson-? inverse analysis load identification for the overall model error.The algorithm can correct most of error of each parameter of the model and has high precision.This method provides a new idea for model correction.