In　order　to　obtain　influence　laws　of　environmental　temperature　on　the　relative　wavelet　entropy　index　SWT,　the　related　study　based　on　health　monitoring　system　of　a　prestressed　concrete　continuous　rigid-frame　bridge　with　box　sections　was　conducted.　Firstly,　the　calculation　method　of　the　relative　wavelet　entropy　index　for　bridge　structures　under　environment　excitation　was　introduced.　Secondly,　the　daily　and　annual　varying　rules　of　the　relative　wavelet　entropy　index　of　each　box　section　of　the　bridge　were　analyzed　and　obtained　based　on　the　real-time　monitoring　data.　Finally,　by　using　the　gradually　optimal　linear　regression　analysis　method,　the　multivariate　linear　regression　model　for　the　relative　wavelet　entropy　index　considering　time　and　spatial　distribution　of　temperature　was　developed,　and　the　temperature　correctional　model　for　the　relative　wavelet　entropy　index　was　proposed.　Results　showed　that　the　multivariate　linear　regression　model　has　a　better　fitting　accuracy　and　a　relatively　high　prediction　precision;　the　effects　of　environmental　temperature　on　the　relative　wavelet　entropy　index　can　be　eliminated　effectively　with　the　temperature　correctional　model　to　make　the　relative　wavelet　entripy　index　be　used　in　bridges＇　damage　identification　effectively.　©　2017,　Editorial　Office　of　Journal　of　Vibration　and　Shock.　All　right　reserved.