在這篇論文中，我們要探討的是關於任意實方陣A如何求出其實部最大的特徵值， 而我們解決此一問題的方法則是透過幾何模型找出複數平面上代表A的特徵值之點(以下簡稱特徵點)中最右側者：首先從A的特徵點中抓出一個位置相對右側者，令之為P1，構造矩陣MT，將A的特徵點經一反演與鏡射變換，使其中在P1左側者變換到某一圓內；P1右側者變換到該圓外。接下來檢驗圓外有無由A的特徵點變換過來之點？若無，則P1 即為所求；若有，則將之逆變換回去，可得A中另一在P1右側的特徵點P2，在矩陣MT的作用下，一樣可以將P2左側的特徵點變換到某一圓內；P2右側的特徵點變換到該圓外。再檢驗圓外有無由A的特徵點變換過來之點？若無，則P2 即為所求；若有，則將之逆變換回去，又可得A中另一在P2右側的特徵點P3，不斷重複迭代上述步驟…，可得一向往右側前進點列＜Pn＞，最後必可得A的特徵點中最右側者。文中並列出幾個實際執行的數值結果來驗證我們的理論。

In this thesis, we find the eigenvalue with maximal real part of a general matrix. We use a geometric model to solve this problem.

[1] C. Moler and C. Van Loan, Nineteen dubious ways to compute the exponential of a matrix,
SLAM Review, (1978), Vol. 20, pp.801 836.
[2] M. Ortiz, R. A. Radovitzky and E. A. Repetto, The computation of the exponential and
logarithmic mappings and their first and second linearizations, Int. J. Numer. Meth. Engng,
(2001), Vol. 52, pp.1431 1441.
[3] Yiguang Liu, Zhisheng You, Liping Cao, A recurrent neural network computing the largest
imaginary or real part of eigenvalues of real matrices, Comp. Math. Appl. (2007), Vol. 53,
pp.41 53.
[4] W. R. Ferng, W.-W. Lin, D. Pierce and C.-S. Wang, Nonequivalence transformation of
-matrix eigenproblems and model embedding approach to model tuning, Num. Lin.
Alg. Appl. (2001), Vol.8(1), pp.53 70.
[5] Axel Ruhe, The Rational Krylov algorithm for nonsymmetric eigenvalue problems. III
Complex shifts for real matrices, BIT, (1994), Vol. 34, pp.1 14.