THE ACQUISITION OF INTERVAL NUMBER FREQUENCY AND ITS APPLICATION TO STRUCTURE DAMAGE DETECTION

摘要

This work aims to enhance the efficacy of structure damage detection based on a natural frequency change approach. The objective is to develop an interval number frequency value, which can be applied to damage detection, by introducing small sample analysis. First, this paper carries out a special study on small sample analysis and proposes the Normality test method for small samples. The interval estimation of the mean value for a small sample, which is supposed to fit Normal distribution, is investigated. The interval number acquisition approach is proposed if the parameters of a small sample are supposed to fit Uniform distribution. Second, this paper provides the acquisition approach of interval number frequency with the combination of the indirect frequency acquisition approach and small sample analysis. This type of interval value is more accurate and reasonable because it can take into account the influence of various uncertainties and errors. Consequently, based on the interval number frequency value, two specific ways in which to apply this method to damage detection are proposed. Finally, the applications and advantages of this approach are illustrated using two examples of actual large bridges in China.