光柵莫爾條紋電子學(xué)細(xì)分技術(shù)研究

《光柵莫爾條紋電子學(xué)細(xì)分技術(shù)研究》由會員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫。

1、南京航空航天大學(xué)碩士學(xué)位論文摘要光柵測量技術(shù)具有高精度、高靈敏性、動態(tài)范圍大、易于實(shí)現(xiàn)自動化等特點(diǎn)，可實(shí)現(xiàn)對位移、速度等機(jī)械量的測量。為提高測量精度和分辨力，有效控制測量成本，莫爾條紋電子學(xué)細(xì)分技術(shù)被廣泛研究和應(yīng)用。本文對莫爾條紋電子學(xué)細(xì)分技術(shù)進(jìn)行了深入研究，提出了提高細(xì)分精度和分辨力的解決方案。從基本原理上歸納總結(jié)實(shí)際應(yīng)用中各種細(xì)分方法的實(shí)現(xiàn)途徑和特點(diǎn)。重點(diǎn)對正切法細(xì)分技術(shù)進(jìn)行研究并對細(xì)分精度影響因素進(jìn)行理論分析；在常規(guī)信號調(diào)理的基礎(chǔ)上，通過算法對莫爾條紋信號進(jìn)行噪聲控制和相位校正，可極大提高細(xì)分精度。為獲得高質(zhì)量信號，

2、將神經(jīng)網(wǎng)絡(luò)自適應(yīng)算法應(yīng)用于莫爾條紋信號降噪。神經(jīng)網(wǎng)絡(luò)層實(shí)現(xiàn)信號的非線性映射，使線性和非線性噪聲均能得到有效抑制；將濾波步長與信號頻率構(gòu)成函數(shù)關(guān)系，步長的動態(tài)調(diào)整和算法的自適應(yīng)性保證了寬頻帶濾波效果。濾波后的莫爾條紋信號質(zhì)量明顯改善?；谡蟹?xì)分，提出了一種新的相位誤差補(bǔ)償算法，通過信號區(qū)間的分段處理，使短周期信號的各相角均可得到實(shí)時(shí)校正，有效降低了由于信號相位不正交引起的細(xì)分誤差，并詳細(xì)分析了算法的實(shí)現(xiàn)條件。仿真及實(shí)驗(yàn)表明，本文提出的算法可顯著改善莫爾條紋信號質(zhì)量；相關(guān)研究內(nèi)容對于提高莫爾條紋電子學(xué)細(xì)分精度和倍數(shù)具有實(shí)際

3、參考價(jià)值。關(guān)鍵詞：光柵，莫爾條紋，細(xì)分，正切法，噪聲，相位I光柵莫爾條紋電子學(xué)細(xì)分技術(shù)研究ABSTRACTGratingMoiréFringetechniqueisusedtomeasurethemechanicalpositionandvelocityatpresentforitshighprecision,perfectsensitivity,widedynamicrange,automatizationandsoon.Inordertoimprovethemeasurementdistinguishmentandpr

4、ecisionanddepresstheapplicationcost,thesubdivisionofMoiréFringeisstudiedandappliedwidely.TheelectricsubdivisiontechniqueofMoiréFringeisstudiedperfectlyandhowtoimprovethesubdivisionprecisionanddistinguishmentareputforwardinthisthesis.Therealizationsandthecharacteri

5、sticsofvarioussubdivisionmethodsaccordingtotheprinciplearesummarizedfirst.ThetangentsubdivisionmethodofMoiréFringeisstudiedwithemphasisandthefactorswhichinfluencethesubdivisionprecisionareanalyzedtheoretically.Toimprovethesubdivisionprecision,thesignalnoisesshould

6、berestrainedandthephaseshouldbeadjustedthroughalgorithm.Inordertoobtainthesignalswithfinequality,theadaptivefilteringalgorithmbasedonneuralnetworkisusedtorestrainnoisesofMoiréFringe.Thenonlinearmappingfuctionisachievedbyusingtheneuralnetworklayer.Thestepsizeofthea

7、lgorithmcanbeadjusteddynamiclyaccordingtothesignal’sfrequencytomeetthefilteringrequestofsignalwithdiversifiedfrequencyandmakethealgorithmself-adaptive.Thesignalqualityisimprovedobviouslybyfiltering.Anewphaseerrorcompensationalgorithmisworkedoutinallusiontothetange

8、ntmethod.Theangleofacertainsignalperiodcanbeemendedbydetachingthesections.Theapplicationofthisalgorithmcandepressthesubdivisionerrorcausedbysignalsnotin