Font Size: a A A

Study Of Damage Identification Based On Modal Parameter

Posted on:2008-05-03Degree:MasterType:Thesis
Country:ChinaCandidate:D B ZhangFull Text:PDF
GTID:2132360212490328Subject:Structural engineering
Abstract/Summary:PDF Full Text Request
Many damage identification methods are analyzed by numerical simulation in this paper. Modified damage identification methods can provide better precision and efficiency. Mode experiments uses two test methods to analyze dynamic modal parameters, in order to provide needs of different identified methods, and using engineering data to analyze modal test.Two-step-method is adopted for damage detection in this paper. Combined using curvature modal parameters and modal assurance criterion is order to identify damage location. According to comparison of many modal parameters, local curvature modal parameters are more sensitive to damage. By analysis of merits and shortages of two test methods, two methods' combination can enhance identified integrity. Modified co-ordination modal assurance criterion can set up uniform standard to select damage degree's identification parameters. BP neural network can identify damage degree. Considering requests of on-line test in structure health monitoring system, damage test model is discussed by using BP neural network.Dynamic test is process of assessment of structure performance, damage and so on by characteristic parameters in identification of structure dynamic response, so it must analyze characters of dynamic modal parameters. Mode experiment uses hammer test and sine frequency sweep to analyze test parameters. It can provide reference standard for selecting test methods. Sensitivity analysis can give some reference of modified mode, frequency analysis can verify relationship between additive mass and frequency change. According bridge modal test programme and analysis results, its can analyze efficiency and shortage of environmental vibration experiment in bridge test.
Keywords/Search Tags:structure health monitoring, damage identification, modal parameter, BP neural network, modal test
PDF Full Text Request
Related items