Finite Element Model’s Updating Of Parameters Based On Tests

Finite element method plays a more and more important role in modern technology. It’s very important to build a reliable model. But the simplification of models causes errors between finite models and experimental models.The objective of model updating is to reduce the errors between finite models and experimental models and make the updated models show the right characteristics and forecast dynamic response. In the paper, a method basing miscorrelation index of finite models and experimental models is used to update the model. First, a row of response functions of experiment model are get through model tests. The miscorrelation index can be obtained by multiplying dynamic stiffness matrix picked up from the file .and response function of experiment model. The equations are usually under determined because modal numbers tested are not enough, and sensitivity matrix may be singular. To solve these problems, additional constraints and masses are added to the model to get different tests’results. The model are updated using only one group of test result, two groups of test results or three groups of test results. Fuzzy evaluation is used to evaluate the updated models,.Garteur is updated using the method above. And fuzzy evaluation is used to evaluate the updated models using only a groups of test results , two groups of test results or several groups of test results. The results of evaluation indicate that updated models using three groups of test results is better than updated models using a group of test results or two groups of test results. The results show that the method using several groups of test results has important engineering value.