小波變換分光光度法同時測定感冒液成分的研究

食品科學論文編號:SP077  論文字數:15353,頁數:41

  
目  錄
 中文摘要........................................................Ⅰ
 英文摘要  ......................................................II
 目   錄  .......................................................III
 1.緒論  .........................................................1
 2.多組分同時測定的原理及實驗方法  ...............................3
 2.1計算分光光度法測定多組分的原理  .........................3
 2.2實驗方法與技術...........................................3
 2.2.1吸收系數矩陣ε的求解................................3
 2.2.2混合標準溶液的配制 .................................4  
 2.2.3檢測波長位置的選擇..................................4
 3.小波變換基本原理...............................................6 
4.實驗部分  .....................................................8
 4.1儀器和主要試劑  .........................................8
 4.2吸收曲線測定.............................................8
 4.3樣品穩定性試驗  .........................................8
 4.4訓練樣本吸光度測定.......................................8
4.5方法適用性試驗...........................................8
4.6樣品加標回收試驗.........................................8
 4.7未知樣品吸光度測定  .....................................8
5.結果與討論.....................................................9
5.1吸收曲線與測定波長的選擇.................................9
5.2樣品穩定性分析...........................................9
5.3訓練樣本的確定 ..........................................10
5.4主成分數的確定 ..........................................12
5.5小波系數的確定 ..........................................12
5.6方法適用性試驗 ..........................................13
5.7加標回收試驗  ...........................................15
5.8未知樣品預報 ............................................15
6.總結與展望 ....................................................17
 致謝     ........................................................18
 參考文獻 ........................................................19
 附錄A 81組分吸光度數據..........................................21
 附錄B MATLAB相關程序 28

摘 要：在多組分樣品的分光光度法測定中，吸收曲線常會發生不同程度的重疊,采用一般分光光度法難于定量每一組分，而近年來發展起來的小波變換現已成為信息科學中的一個新興研究領域，越來越多地應用于化學研究領域中各種復雜的、非線性的問題，也取得了很好效果。本文采用原始吸光度數據的1尺度低頻系數C構造分析矩陣進行PLS回歸分析，預測結果優于原始數據全譜模型。本文用小波變換-偏最小二乘回歸分光光度法對對乙酰氨基酚(ACET)、愈創木酚甘油醚（GUAI）、咖啡因（CAFF）和撲爾敏（CHLO）的四組分體系進行了同時測定。加標回收率介于100.36-118.50%，結果基本滿意。使用小波系數進行PLS建模，不僅可以消除原始信號中的部分噪聲，而且較大地壓縮了數據量，提高了預測精度。綜上所述，小波變換-偏最小二乘法結合分光光度法為多組分的同時測定提供了一種新的簡便、可靠的方法。

關鍵詞：小波變換；偏最小二乘；紫外分光光度法；多組分同時測定；感冒液

Abstract: Determinate the multi-component samples, the absorption curve offen overlaps in varying degrees , It is difficult to quantitate each component using spectrophotometry generally, but the wavelet transforrrmtion (WT) in the field of chemical application which particularly in the field of nonlinear applications are very wide and have achieved good results. In this paper,Using a standard low-frequency coefficient C1 for the original absorbance data to structure analysis matrix to PLS regression analysis, forecasting results is better than the original entire spectrum model.In this paper using WT-PLS spectrophotometry to simultaneously determinate  the Cough Syrup which has four components: acetaminophen, guuaifenesin, caffeine, p-aminophenol.the recovery ranged from 100.36% to 118.50% ,the result is basically satisfied. Using wavelet coefficients to build PLS model, not only can eliminate some of noise in the original signal , but also can greatly reduce the volume of data,the forecast accuracy can be improved,.In conclusion,the WT-PLS spectrophotometry provides a simple, reliable method for the simultaneous determination of multi-component.

Keywords: wavelet transforrrmtion；PLS；UV spectrophotometry；simultaneous determination of multiple components；Cough Syrup