Font Size: a A A

A Research On Moving Load Identification Based On Structure Response Reconstruction

Posted on:2013-11-09Degree:MasterType:Thesis
Country:ChinaCandidate:L ChiFull Text:PDF
GTID:2232330374490681Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
In practical engineering problems, the form of the loads acted on the structure ismoving dynamic load in many cases, and the moving load identification for largestructure and complex system has important significance.A research on moving load identification based on structure responsereconstruction is discussed. In the moving load identification process, because themoving load action position is varying along with the time transformation, a lot ofstructural response information is needed. However, the measure point on structure isalways limited; we cannot get the entire structural response. At the same time, becauseof the moving load identification inverse problem is often ill-posed, even if a smallnoise was contained in the measured response, it would also be magnified seriously bythe sick system matrix, leading the serious deviating between reverse results and thepractice. To solve these problems, the moving load calculation reverse technologybased on dynamic response signal reconstruction is discussed. The analysis isdeveloped around three aspects of the dynamic equation, dynamic moving load, systemmatrix and the dynamic response of the structure. Using the dynamic response fromthe limited measure point of structure, the whole structure dynamic response signal isreconstructed. After getting reconstruction response, the moving load was obtained viathe dynamic equation. The main research work of this paper was:(1) Deconvolution dynamic load identification algorithm based on structureresponse signal reconstruction is analyzed. Radio Base Function (RBF) is used as thebase function, the response signal is reconstructed. Based on system modedecomposition, the system structure response and system dynamic load was projectedto the modal coordinates. Using modal coordinates decoupling property, the Greenfunction is got, the convolution method to calculate the moving load in modal domainis used, and project it back to the physical coordinate system to get the moving load.(2) Central difference dynamic load identification algorithm based on structureresponse signal reconstruction is analyzed. For engineering problem, in order to finishmoving load identification mode selection, load identification method based on centraldifference was adopted. The discomfort of the method and the time interval wasdiscussed.(3) Dynamic load identification algorithm based on prior knowledge is analyzed. Inthe process of moving load identification there are a lot of influences fromenvironment and practical condition. Inversed model may not be particular, and theresponse information measured is usually not complete. If some prior knowledge aboutthe moving load is known in advance, the scale of the inverse project can be reduced, and the reverse illposedness would be reduced. The identification with known area andknown forms of moving load is mainly discussed in this paper. Via separation of thepossible structure subsystem, the signal quantity would be reduced, In this way, theresult of the inverse would be more accurate and stable.(4) Dynamic load identification algorithm based on interval analysis of uncertainstructural parameters is analyzed. In the actual engineering, the complexity of thestructure and material attribute uncertainty would cause the mode uncertainty whenbuild the mode of the moving load identification. In order to get an objective result ofmoving load identification, based on the uncertain system’s dynamic equation, usingthe theory of interval mathematics, the uncertain structure moving load identificationwas translated into two kinds certain inverse problem which would be solved bycertain structure system method to make objective assessment to the moving load onthe structure.
Keywords/Search Tags:Moving load identification, Inverse problem, Response reconstruction, Prior knowledge, Uncertainty
PDF Full Text Request
Related items