Analysis Based On Least Squares Support Vector Machine Under Deck Tied Arch Bridge Assessment

With the development of the high way, the growing of traffic volume and the increasing of heavy vehicle, the exiting bridge loading capacity is more and more urgent to be assetted. In present the methods to asset the brige loading capacity are based on the in situ static and dynamic test with the high cost.In this paper a new kind of bridge evaluation method based on support vector machine is proposed, an support vector machine (SVM) assetment model was established based on the existing bridge bearing capacity test data.with this model the loading capcatiy of a bridge can be estimated by the inspection results.(1) Take the Yellowstone bridge in Wuxi city as engineering background, the present methodes to asset the loading capacity based on static and dynamic load were evaluated.(2) Using the modification of the appearance inspection (MAI) methods, the Yellowstone bridge bearing capacity was evaluated. By comparing MAI methods and the in situ static and dynamic test results, feasibility of appearance inspection is verified and and the inspection results can be selected as the support vector.(3) For the detection of six concrete-filled arc bridge support vector selection, make it as the support vector machine (SVM) model of the sample, will be to evaluate bridge support vector as the system input, the evaluation results as the system output, the example of application of support vector machine (SVM), to validate the feasibility in the actual bridge examination system.