The　active　structural　health　monitoring　(SHM)　approach　for　the　complex　composite　laminate　structures　of　wind　turbine　blades　(WTBs),　addresses　the　important　and　complicated　problem　of　signal　noise.　After　illustrating　the　wind　energy　industry＇s　development　perspectives　and　its　crucial　requirement　for　SHM,　an　improved　redundant　second　generation　wavelet　transform　(IRSGWT)　pre-processing　algorithm　based　on　neighboring　coefficients　is　introduced　for　feeble　signal　denoising.　The　method　can　avoid　the　drawbacks　of　conventional　wavelet　methods　that　lose　information　in　transforms　and　the　shortcomings　of　redundant　second　generation　wavelet　(RSGWT)　denoising　that　can　lead　to　error　propagation.　For　large　scale　WTB　composites,　how　to　minimize　the　number　of　sensors　while　ensuring　accuracy　is　also　a　key　issue.　A　sparse　sensor　array　optimization　of　composites　for　WTB　applications　is　proposed　that　can　reduce　the　number　of　transducers　that　must　be　used.　Compared　to　a　full　sixteen　transducer　array,　the　optimized　eight　transducer　configuration　displays　better　accuracy　in　identifying　the　correct　position　of　simulated　damage　(mass　of　load)　on　composite　laminates　with　anisotropic　characteristics　than　a　non-optimized　array.　It　can　help　to　guarantee　more　flexible　and　qualified　monitoring　of　the　areas　that　more　frequently　suffer　damage.　The　proposed　methods　are　verified　experimentally　on　specimens　of　carbon　fiber　reinforced　resin　composite　laminates.