Damage Identification Of Random Structures Using Static And Dynamic Measurement Data

With the rapid development of all kinds of dynamic and static testing techniques as well as their corresponding techniques of digital signal processing, the research of structural nondestructive evaluation has been promoted greatly, and many damage identification methods based on structural dynamic or static parameters have sprung up. On the basis of previous studies, this paper explores the problem of damage identification of random structures influenced by model error and measurement noise using structural static and dynamic measurement data. And the following research results are obtained:1.By means of stochastic finite element method, a damage identification method of random structures based on structural static deflection measurement is developed. Using the method, the statistical characteristics of structural random damage index of each element can be obtained, and then the element's probability of damage existence can be assessed. The numerial examples show that the identification results of the method are very close to those of Monte-Carlo simulation.2. A damage identification method for random structures based on static and dynamic combined measurement data is proposed in this paper. According to this method, dynamic measurement is used to revise the static identification results adjusted after probabilistic analysis. The numerical examples show that with model error and measurement noise taken into consideration, this method is effective in identifying damage cases of random beam structures using static deflection measurement and less dynamic modal measurement, and is of higher precision than the method based on either type of the above measurement data under the same conditions.3. An improved damage identification method for random structures based on static measurement data is presented. This method locates the undamaged elements on the basis of probabilistic analysis of the static identification results, which then are adjusted and substituted into the static governing equations to re-identify structural damage. The numerical examples show that the method proposed can identify damage of random beam structures effectively. Moreover, It can lower the risk of misjudgments, and improve the accuracy of static damage identification. Also, it is found that the stiffness variation of an actual structure can cover its real damage, which may bring down identification accuracy of the method proposed.