English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 49064/83170 (59%)
造訪人次 : 6961195      線上人數 : 36
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/111491


    題名: 結合子空間演算法及隨機式全域最佳化演算法重建二維介質物體
    其他題名: Dielectric objects reconstruction by combining subspace-based algorithm with randomly global optimization algorithm
    作者: 顏健佑;Yen, Chien-Yu
    貢獻者: 淡江大學電機工程學系碩士班
    丘建青
    關鍵詞: 微波成像;子空間演算法;全域演算法;自我適應之差異型演化法;Microwave imaging;Subspace-based Algorithm;Global Optimal Algorithm;Self-Adaptive Dynamic Differential Evolution (SADDE)
    日期: 2016
    上傳時間: 2017-08-24 23:54:06 (UTC+8)
    摘要: 本論文探討子空間演算法(Subspace-based algorithm)應用於自由空間中二維介質物體之逆散射問題。處理逆散射問題的方法中,子空間演算法特別不同的地方在於計算上使用到奇異值分解(Singular value decomposition , SVD),運用子空間的概念,將感應電流分成確定性部分及不確定性部分,確定性部分對於逆散射提供良好初始猜測值,逆散射只針對不確定性部分做運算及最佳化,這部分是子空間演算法的精華,可以在計算上減少未知數的數量,有效降低計算成本及簡化計算過程。最佳化演算法方面再使用自我適應之動態差異型演化法(Self-Adaptive Dynamic Differential Evolution, SADDE),避免像使用共軛梯度法(Conjugate Gradient method, CG)或牛頓法(Newton''s Method)會容易陷入區域極值的問題,雖然避免了區域極值的問題,但計算時間卻會增加,因此利用子空間演算法本身簡化計算降低成本的優點再配合SADDE之強健性和搜尋速度,收斂至更佳的結果,並增加對雜訊的抗性。此外同時比較子空間演算法分別搭配SADDE和基因演算法(Genetic Algorithm, GA)之結果顯示子空間演算法在演算法方面搭配SADDE有較佳的重建結果。另外進一步討論子空間演算法對於複雜非均勻介電物體的重建及對於雜訊的優良抗性,研究模擬之數值結果顯示,此數值方法運用於重建複雜非均勻介電物體之材質特性分佈,皆能得到良好之重建結果,且無論加入雜訊等級的大小,儘管已使數據與正確值相差甚大,皆能藉由子空間演算法之參數調整,收斂至更良好之重建結果。
    This thesis presents the two-dimensional electromagnetic imaging problem by Subspace-based algorithm. Subspace-based algorithm is different with methods of processing inverse scattering problem by contrast source inversion (CSI). The essence of the subspace-based optimization method is that part of the contrast source is determined from the spectrum analysis without using any optimization when the rest is determined by optimization method. By applying the singular value decomposition (SVD) to the field equation, the induced current is divided into the signal space and the noise space. This feature can reduce the number of unknowns and computing costs to speed up the convergence of the algorithm. We also transform the inverse scattering problem into optimization problem and solved by Self-Adaptive Dynamic Differential Evolution (SADDE). SADDE can process numerous unknowns of electromagnetic imaging problems. Different scatterers and environment will be used to investigate whether Subspace-based algorithm can keep stability of reconstruction or not. We will also compare Genetic Algorithm (GA) to show the robustness and the searching speed of SADDE.
    顯示於類別:[電機工程學系暨研究所] 學位論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    index.html0KbHTML1檢視/開啟

    在機構典藏中所有的資料項目都受到原著作權保護.

    TAIR相關文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回饋