The　cable　prestressing　loss　and　web　members　damage　experiment　of　a　beam　string　structure　was　carried　out.　Different　prestressing　forces　were　exerted　on　the　cable　to　simulate　cable　prestressing　loss,　the　prestressing　forces　were　calculated　by　the　frequencies　which　were　solved　by　Fourier　transform　or　wavelet　transform　under　impulse　excitation.　Test　results　show　this　method　is　accurate　and　convenient.　The　damage　cases　of　web　members　on　the　beam　were　tested　to　validate　the　efficiency　of　wavelet　support　vector　machine　method.　Wavelet　packet　decomposition　was　applied　to　the　structural　response　signals　under　ambient　vibration,　feature　vectors　were　obtained　by　feature　extraction　method　with　＇energy-damage　state＇.　The　feature　vectors　were　used　for　training　and　classification　as　the　inputs　of　the　compact　support　vector　machine.　The　structural　damage　position　and　degree　can　be　identified　and　classified,　and　the　test　result　is　highly　accurate　especially　combined　with　principle　component　analysis.