341 期末考的範圍遠遠多於期中考,要了解的定理和觀念也非常多 希望同學們能及早準備 342 Chapter 6 Series Solutions of Linear Equations 假設 DE 的 solutions 為 polynomial 的型態 (和 Cauchy-Euler Method 以及 Taylor Series 的概念相近) yx cn x x0 n n0 被稱作為 power series centered at x0 Power series 觀念的複習 (Sec. 6-1) x0 is a non-singular point (Sec. 6-2) regular singular point (Sec. 6-3) x0 is a singular point Example (Sec. 6-4) irregular singular point 343 Section 6-1 Reviews of Power Series 6-1-1 定義 1. Power series c n x x 0 c 0 c1 x x 0 c 2 x x 0 n 2 n0 N 2. Convergence: lim N c x x n 0 n exists n0 測試方法:Ratio test (test for convergence) lim n c n 1 x x 0 cn x x0 n 1 n L L < 1: converge L = 1: 不一定 3. Radius of Convergence R L < 1 if |x − x0| < R L > 1 if |x − x0| > R L > 1: diverge 344 Example 1 (text page 233) For the Power series n 1 x 3 lim 2 n 1 n n 2 n n 1 ( n 1) x 3 n x 3 n x 3 lim n x3 n 2( n 1) 2 n 2 n x3 2 1 for 1 < x < 5 Interval of convergence: (1, 5) However, since when x = 1, the power series becomes 1 / n , n n 1 which is also convergent, the interval of convergence is modified as: Interval of convergence: [1, 5) 345 6-1-2 Maclaurin Series (Taylor Series) y x y x0 y (4) y x0 x0 4! 1! ( x x0 ) ( x x0 ) 4 y x 0 2! ( x x0 ) 2 y x 0 3! ( x x0 ) 3 346 Maclaurin Series (Taylor Series) 2 3 (-∞, ∞) e 1 x x x 1! 2 ! 3! x 3 5 2 4 sin x x x x 3! 5! (-∞, ∞) 6 cos x 1 x x x 2! 4! 6! 2 Interval of Convergence 3 (-∞, ∞) 4 ln 1 x x x x x 2 3 4 (-1, 1] 1 1 x x2 x3 1 x (-1, 1) e 2x ? cos 3 x ? 347 Example 2 (text page 235) Find a power series representation of exsinx 2 3 4 e sin x 1 x x x x 2 6 24 x 3 5 x x x 6 120 2 3 4 5 (1) x (1) x 1 1 x 1 1 x 1 1 1 x 6 2 6 6 120 12 24 2 3 5 xx 1x 1 x 3 30 Section 6-2 Solutions about Ordinary Points c x x Suppose that the solution is n n 0 n0 6-2-1 方法適用情形 (1) Linear (n) ( n 1) a n x y a n 1 x y a1 x y a 0 x y g x (2) x0 is not a singular point (3) It is better that a0(x), a1(x), …., an(x), g(x) are all polynomials. (or expressed as a Taylor series) 348 349 6-2-2 解法流程 Step 1 將 y x c n ( x x0 ) n 代入 (x0 必需為 ordinary point) n0 Step 2 對齊 (一律變成 (x - x0)k ) 不是 singular point 的 即為 ordinary point Step 3 合併 Step 4 比較係數,將 cn 之間的關係找出來 Step 5 Obtained independent solutions and general solution 350 6-2-3 例子 Example 5 (text page 241) y xy 0 Set yx cn x n since P(x) = 0 and Q(x) = x are analytic at 0 n0 Step 1 y xy c n n ( n 1) x n2 n2 c n n ( n 1) x x cn x 0 n n0 n2 n2 cn x set k = n + 1 c k 2 ( k 1) x k 2 k 0 0 n0 set k = n −2 Step 2 對齊 n 1 k c k 1 x 0 k k 1 c k 2 ( k 1) x 351 k k 2 k 0 c x 0 k k 1 k 1 Step 3 2 c 2 [ c k 2 k 2 ( k 1) c k 1 ] x k 0 k 1 Step 4 c k 2 k 2 ( k 1) c k 1 0 2c2 = 0 c k 1 或 c k 2 k 2 ( k 1) recurrence relation c0, c1 給定之後 c2 = 0 k=1 k=2 k=3 c0 2 3 c1 c4 34 c c5 2 0 4 5 c3 ck 2 k=4 k=5 k=6 k=7 k=8 k=9 c k 1 k 2 ( k 1) c3 c0 c6 5 6 2 35 6 以此類推,所有的 cn 的值都 可以算出來 (以 c0 或 c1 表示) c4 c1 c7 67 3467 c c8 5 0 7 8 c6 c0 c9 89 2 3 5 6 8 9 c7 c1 9 10 3 4 6 7 9 10 c8 0 10 11 c10 c11 : : 352 Step 5 353 x x y x c n x c 0 1 x 2 3 2 3 5 6 2 3 5 6 8 9 n0 3 n 6 9 x x c1 x x 3 4 3 4 6 7 3 4 6 7 9 10 4 7 y x c 0 y 1 x c1 y 2 x y1 x 1 2 3 k 1 y2 x x 34 k 1 ( 1) 3k x (3 k 1)(3 k ) k ( 1) 3 k 1 x (3 k )(3 k 1) k 10 y1 y 2 354 Example 6 (text page 243) yx ( x 1) y xy y 0 2 cn x (analytic at x = 0) n n0 Radius of convergence? Step 1 x 1 n n 1 cn x n2 2 n n 1 cn x n n2 Step 2 k 2 x nc n x n 1 n n 1 cn x n2 n2 n 1 k k 1 ck x k cn x 0 n n0 nc n x n n 1 cn x 0 n n0 k=n−2 k=n n2 k=n k=n k k 2 k 1 c x kc x c x k k 0 k 2 k k 0 k k 1 k 0 k k 1 c x k k k 2 k 2 k 1 c k 0 x k k 2 kc 355 x k k k 1 c x 0 k k k 0 Step 3 2 c 2 c 0 (6 c 3 c1 c1 ) x k=0 k=1 + [ k k 1 c k k 2 k 1 c k 2 kc k c k ] x 0 k k 2 2 c 2 c 0 6 c 3 x + [( k + 1) k 1 c k k 2 k 1 c k 2 ] x 0 k k 2 Step 4 2 c 2 c 0 =0 6 c 3 =0 ck 2 = k 1 ck k2 2 c 2 c 0 =0 6 c 3 =0 c 2 = c 0 /2 c 3 =0 ck 2 = 1 k ck k2 356 c0, c1 給定之後 c 2 = c 0 /2 c3 = 0 c 4 1 c 2 1 c 0 21 c 0 4 24 2 2! c5 2 c3 0 5 3 c 3 c c6 3 c4 6 2 4 6 0 2 3 3! 0 c7 4 c5 0 7 c 8 5 c 6 3 5 c 0 34 5 c 0 8 2 4 6 8 2 4! c9 6 c7 0 9 : : 357 Step 5 y x c n x c 0 1 x 2x 33 x 3 4 5 x 3 55 7 x 2 2 2 ! 2 3! 2 4 ! 2 5! n0 2 n 4 6 2 y1 x 1 1 x 2 y2 x x n2 ( 1) 10 c x 1 y1 y x c 0 y1 x c1 y 2 x 8 n 1 13 5 (2 n 3) n 2 n! x 2n |x| < 1 (Why?) y2 358 Example 8 (text page 245) y (co s x ) y 0 2 4 6 cos x 1 x x x 2! 4! 6! 2 4 y1 1 1 x 1 x 2 12 3 5 y2 x 1 x 1 x 6 30 6-2-4 定義 1. Analytic at x0: If a function can be expressed as a power series and the radius of convergence of the power series is nonzero 簡單的判斷 f(x) 在 x0 是否為analytic 方法 (1) f(x0) should be neither nor − (2) f(m)(x0) should be neither nor − m = 1, 2, 3, ………. 359 360 2. Ordinary Point and Singular Point: For the 2nd order linear DE a 2 x y a1 x y a 0 x y 0 y P x y Q x y 0 Definition 6.1 x0 is an ordinary point of the 2nd order linear DE if both P(x) and Q(x) are analytic at x0 Otherwise, x0 is a singular point . Theorem 6.1 If x0 is an ordinary point of the 2nd order linear DE, then we can find two linearly independent solutions in the form of a power series centered at x0 , i.e., yx n0 cn x x0 n For the kth order linear DE ak x y y (k ) 361 (k ) a k 1 x y Pk 1 x y ( k 1) ( k 1) a1 x y a 0 x y 0 P1 x y P0 x y 0 Pm x Extension of Definition 6.2.1 am x an x x0 is an ordinary point of the kth order linear DE if P0(x), P1(x), P2(x), ………. , Pk−1 (x), are analytic at x0 Otherwise, x0 is a singular point . Extension of Theorem 6.2.1 If x0 is an ordinary point of the nth order linear DE, then we can find n linearly independent solutions in the form of a power series centered at x0 , i.e., yx n0 cn x x0 n 6-2-5 Interval of Convergence 的判斷方法 判斷方法一: 找出 lim n c n 1 ( x x 0 ) cn ( x x0 ) n 1 n 1 的條件 判斷方法二 (較快速,但較不精準 找出的收斂的範圍有時會比實際的收斂範圍小) x x0 R 其中 R 是 x0 和最近的 singular point 的距離 Singular point can be a complex number , see the example on page 354 超過這個範圍未必不為 convergence 362 363 6-2-6 思考 (1) 對於 nonhomogeneous 的情形….. an x y (n) x a n 1 x y ( n 1) x (2) 這方法還可以用在什麼情形? a1 x y x a 0 x y x g x 6-2-7 本節需注意的地方 364 (1) 要了解幾個重要定義: (a) convergence, (b) radius of convergence, (c) analytic at x0, (d) singular point, (e) ordinary point (2) 複習一下 Taylor series (如 page 346) (3) Index 的地方計算要小心 (a) 先都化成 xk 再合併,(b) 頭幾項可能要獨立出來 (c) Index 對齊計算要小心 (4) nth order linear DE 要有 n 個 linearly independent 解 (5) 有時要考慮 interval of convergence Section 6-3 Solutions about Singular Points y (n) Pn 1 x y 假設解為 ( n 1) P1 x y P0 x y 0 yx cn x x0 nr n0 6-3-1 方法適用情形 (1) Linear (2) (x −x0)Pn−1 (x), (x −x0)2Pn−2 (x), …………. , (x −x0)n−1P1(x), (x −x0)nP0(x) are analytic at x0 (比較: Section 6-2 要求 Pn−1 (x), Pn−2 (x), …………. , P1(x), P0(x) are analytic at x0) (3) It is better that P0(x), P1(x), …., Pn-1(x) are all polynomials. 365 366 6-3-2 定義 Singular Points 分成二種 If x0 is a singular point but (x −x0)Pn−1 (x), (x −x0)2Pn−2 (x), …………. , (x −x0)n−1P1(x), (x −x0)nP0(x) are analytic at x0 x0 : regular singular point If (x −x0)Pn−1 (x), (x −x0)2Pn−2 (x), …………. , (x −x0)n−1P1(x), (x −x0)nP0(x) are not analytic at x0 x0 : irregular singular point 367 Example 1 (text page 248) ( x 4 ) y 3( x 2 ) y 5 y 0 2 P x 2 3 2 ( x 2 )( x 2 ) Q x 5 2 2 ( x 2) ( x 2) x = 2 is a point x = −2 is a point 368 6-3-3 解法 解法的關鍵: 假設解為 yx cn x x0 nr n0 Theorem 6.3.1 Frobenius’ Theorem 若 x0 是 linear DE 當中的一個 regular singular point 則這個 linear DE 至少有一個解是 yx n0 cn x x0 nr 的型態 369 Process Step 1 將 yx nr c n x x 0 代入 n0 Step 2 Power 對齊 Step 3 合併 Step 4 算出 r Step 5 比較係數,將 cn 之間的關係找出來 Step 6 將 Step 4 得出的 r 代入 Step 5 得出所有的 independent solutions 及 general solution Step 7 (見後頁) 370 (Step 7) 當 (1) r 有重根 或 (2) r 的根之間的差為整數,且從 Step 6 得出來的解不為 independent 時 e P ( x ) dx 用 y 2 x y1 x y 2 1 x dx 和長除法找出 y2(x) (參考 Section 6-3 的 Examples 4, 5) 當 r 的根之間的差為整數,但從 Step 6 得出來的解為 independent 時, 不需進行這個步驟 371 6-3-4 範例 Example 2 (text page 250) 3 xy y y 0 Step 1 將 yx cn x nr 代入 n0 3 c n ( n r )( n r 1) x n r 1 n0 Step 2 Power 對齊 n (n r ) x n r 1 n0 n=k 3 c k ( k r )( k r 1) x k 0 c c n c k 0 nr 0 n0 n=k−1 k=n+1 n=k k r 1 x (k r ) x k k r 1 c k 1 x k 1 k r 1 0 372 Step 3 合併 3 c 0 r ( r 1) c 0 r x r 1 k r 1 3 c ( k r )( k r 1) c ( k r ) c x 0 k k k 1 k 1 Step 4 算出 r 3 r ( r 1) r 0 r 0 Step 5 3r 2 r 0 2 or r (3 r 2) 0 2/3 3 c k ( k r )( k r 1) c k ( k r ) c k 1 0 ck 1 c k 1 ( k r )(3 k 3 r 2 ) Step 6 當r=0 c k 1 ck k (3 k 2) ck 當 r = 2/3 c1 c0 c2 3 7 3!1 4 7 c3 c0 c4 4 10 4 !1 4 7 10 c0 c2 c3 11 3 3!5 8 11 c4 : cn c3 c0 14 4 4 !5 8 11 14 : c0 n !1 4 7 1 c (3 k 2 ) k k 1 c0 5 1 c0 c c2 1 8 2 2 !5 8 c0 c1 1 1 c0 c c2 1 2 4 2 !1 4 c3 373 1 ck c k 1 ( k r )(3 k 3 r 2 ) (3 n 2 ) cn c0 n !5 8 11 (3 n 2) 374 Solution of Example 2 (別忘了將最後的解寫出) y C 1 y1 C 2 y 2 y1 x x 1 0 y2 x x 2 /3 n !1 4 7 n x (3 n 2) 1 n 1 1 n !5 8 11 n 1 1 x (3 n 2) n x 0, (別忘了寫出 x 的範圍) Examples 4, 5 375 (text pages 253, 254) xy y 0 yx Step 1 將 cn x nr 代入 n0 c n ( n r )( n r 1) x n r 1 n0 nr 0 n0 Step 2 對齊 n = k −1 n=k cn x c k ( k r )( k r 1) x k 0 k r 1 c k 1 x k r 1 0 k 1 Step 3 合併 c 0 r ( r 1) x r 1 [c k 1 k ( k r )( k r 1) c k 1 ] x k r 1 0 Step 4 376 r ( r 1) 0 r 0 or 1 c k 1 Step 5 ck Step 6 當 r = 1 ck ( k r )( k r 1) c n ( 1) c k 1 ( k 1) k c0 n ( n 1)! n ! y1 x x 1 n0 n 1 ( 1) n 1 x n !( n 1) ! n ( 1) ( 1) n n x x x n !( n 1)! n 0 n !( n 1)! n n Step 6 ck 當r=0 377 c k 1 ( k 1) k k = 1 時不能算 此時,應該根據 Step 3,由 c k ( k r )( k r 1) c k 1 c1 0 c 0 0 (k = 1, r = 0 代入) c0 必需等於 0, c1 可為任意值 c2 cn c1 c3 1 2 n ( 1) c1 c2 2 3 c1 1 2 2 3 …………. 這地方容易犯錯, 要小心 ( n 1)! n ! y2 x x x 0 n 1 ( 1) n ( n 1)! n ! x n2 n 1 ( 1) n x ( n 1)! n ! n 1 ( 1) m 1 x m !( m 1) ! m 0 m=n−1 m 378 因為前頁算出來的 y2(x) 等於 y1(x), 只好另外用 Sec. 4-2“reduction of order” 的方法求解 e P ( x ) dx y 2 x y1 x y 2 1 x dx e 379 0 dx y 2 x y1 x y1 x y1 x 2 dx y1 x dx 2 3 4 [x 1 x 1 x 1 x 2 12 144 ] 2 dx 2 3 4 5 [x x 5 x 7 x 12 72 y1 x 12 1 7 19 x x x 12 72 long division ] 長除法 dx 2 y1 x 1 ln x 7 x 19 x x 12 14 4 2 y1 x ln x y1 x 1 7 x 19 x x 12 144 2 y1 x l n x 1 1 x 1 x 2 2 思考:為何不是 ln|x| ? 380 6-3-5 多項式的長除法 計算 1 2 3 4 5 x x 5 x 7 x 12 72 (x -2) (x -1) (1) (x) 1 1 (x2) 1 5 12 (x3) (x4) 7 72 (x5) 1 7 12 19 72 1 5 12 7 72 1 5 12 1 1 7 72 5 7 12 72 1 1 7 12 7 12 23 72 7 12 19 72 35 144 49 864 381 6-3-6 Indicial Equation 2nd order case y P x y Q x y 0 If x0 is a regular singular point ( x x 0 ) y ( x x 0 ) p x y q x y 0 2 where p x ( x x 0 ) P x q x ( x x0 ) Q x 2 由於 p(x) 和 q(x) 皆為 analytic p x a 0 a1 x x 0 a 2 x x 0 2 q x b0 b1 x x 0 b 2 x x 0 2 yx cn x x0 nr , y x n0 cn ( n r ) x x0 n r 1 , n0 y x n0 c n ( n r )( n r 1) x x 0 nr2 , 382 將 y(x), y'(x), y''(x), p(x), q(x) 代入 2 ( x x 0 ) y ( x x 0 ) p x y q x y 0 c n ( n r )( n r 1) x x 0 nr n0 a 0 a1 x x 0 a 2 x x 0 2 cn ( n r ) x x0 n0 b0 b1 x x 0 b 2 x x 0 2 cn x x0 nr 0 n0 其中 (x x0) r 的 coefficient 為 c 0 r ( r 1) c 0 a 0 r c 0 b0 r ( r 1) a 0 r b0 0 indicial equation nr yx 383 cn x x0 nr n0 當 linear DE 為 2nd order 時,r 可以由 r ( r 1) a 0 r b0 0 求出 其中 a0 = p(x0) p x ( x x0 ) P x b0 = q(x0) q x ( x x0 ) Q x 2 y P x y Q x y 0 384 For the 2nd order case r ( r 1) a 0 r b0 0 two roots: r1, r2 (Case 1) r1 r2 and r1, r2 are real, r2 − r1 integer 可以找出兩組 y x cn x x0 nr 的解 n0 (Case 2) r1 r2 and r1, r2 are real, r2 − r1 = integer 一個解是 y1 x cn x x0 n0 n r1 另一個解是 y 2 x C y1 x ln x b n x x 0 n r2 n0 C 有時等於 0 (和 case 1 相同) 有時不為 0 385 (Case 3) r1 = r2 時 y1 x cn x x0 n r1 n0 y 2 x C y1 x ln x 或寫成 y 2 x y1 x ln x b x x n 0 C 一定不為 0 n0 bn x x 0 n0 (Case 4) r1 r2 and r1, r2 are complex 在此不予討論 n r2 n r2 bn bn / C 6-3-7 Indicial Equation for Higher Order Case (補充) 386 當 linear DE 為 nth order 時 y (n) Pn 1 x y yx ( n 1) c x x n P1 x y P0 x y 0 0 nr 當中的 r 可以由 n0 r! a r! r! a n 1,0 n 2 ,0 ( r n )! ( r n 1)! ( r n 2 )! a1,0 求出 其中 a k ,0 p k ( x 0 ), p k ( x0 ) ( x x0 ) nk Pk ( x 0 ) k = 0, 1, 2, …., n −1 r! a 0 0 ,0 ( r 1)! 6-3-8 本節需要注意的地方 (1) Index 「對齊」的計算要小心 (建議可以向 power 較小的對齊,否則會出現負的 k) 例如在 page 371 的 Step 2,一律對齊為 xk+r– 1 而非 xk+r (2) 若 x = 0 為 regular singular point, 設 x0 = 0 即可 (3) 如果是 ck 和 ck−1 (或 ck−1 和 ck) 的 recursive relation 其實有時可以立刻將 cn 的式子觀察出來 (但是分母不可變為 0) (4) 小心分母為 0 的情形 (如 page 377) (5) 小心算出來的 y2(x) 和 y1(x) 相同的情形 (如 pages 377, 378) 387 388 (6)別忘了將最後的解寫出 最後的解易出錯的地方: 別忘了乘上 xr (7) Interval of solution 依然要考慮, 且 interval 不包括任何singular point, 即使是 regular singular point (8) 複習長除法 Section 6-4 Special Functions (本節這學期只教不考) Special cases of Sections 6-2 and 6-3 Bessel’s equation of order v x y xy ( x v ) y 0 2 2 Jv x n0 Yv x 2 ( 1) x n ! 1 v n 2 n Solution: c1 J v x c 2Y v x 2nv co s v J v x J v x sin v : 1st kind Bessel function : 2nd kind Bessel function Legendre’s equation of order n (1 x ) y 2 xy n ( n 1) y 0 2 One of the solution: Legendre polynomials (See page 406) 389 390 其他名詞 Gamma function x t x 1 t e dt 0 modified Bessel equation of order v 2 2 2 解:c1Iv( x) + c2Kv(x) x y xy ( x v ) y 0 modified Bessel equation of the 1st kind modified Bessel equation of the 2nd kind Iv x i Kv x Bessel 的另一種變型 x y (1 2 a ) xy ( b c x 2 2 2 2c v J v ix Iv x Iv x sin v 2 a p c )y 0 解 2 2 2 a c c y x c1 J p ( bx ) c 2Y p ( bx ) spherical Bessel functions: Jv(x), v 1 , 3 , 5 , 2 2 2 391 6.4.1 Bessel’s Equation 6.4.1.1 Solving for Bessel’s equation of order v x y xy ( x v ) y 0 2 2 Steps 1~3 將 y x 2 cn x nr 代入 n0 經過一些計算 (詳見text page 257) 得出 c 0 ( r v ) x c1 ((1 r ) v ) x 2 2 r 2 2 r 1 [ c k (( k r ) v ) c k 2 ] x 2 k 2 two roots: v and −v Step 4 r 2 v 2 0 Step 5 c1 (1 r ) v 2 c1 0 2 0 ck ck 2 v (k r ) 2 2 2 rk 0 ck ck Step 6 當 r = v ck 2 392 v (k r ) 2 ck 2 2 當 r = −v k (k 2v) ck ck 2 k (k 2v) 由於 c1 = 0, c3 = c5 = c7 = c9 = ….. = 0 c 2 n ( 1) n 246 c0 2 n (2 2 v )(4 2 v )(6 2 v ) ( 1) c 0 (2n 2v) n 2 2n n !(1 v )(2 v )(3 v ) c 2 n ( 1) n 246 when r = v (n v) c0 2 n (2 2 v )(4 2 v )(6 2 v ) ( 1) c 0 (2n 2v) n 2 2n n !(1 v )(2 v )(3 v ) (n v) when r = −v 6.4.1.2 Gamma function: a generalization of n! x t x 1 t e dt 0 properties of Gamma function (1) n 1 n ! when n is a positive integer 1 0 ! 1 (2) x 1 x x 參照課本 Appendix 1 393 394 (3) n when n is a negative integer or n = 0 (4) 12 1 2 2 6 (x) 4 2 0 -2 -4 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 x 395 6.4.1.1 回到 Solving for Bessel function ( 1) c 0 n c2 n Set c0 c2 n 2 2n n !(1 v )(2 v )(3 v ) 1 v 2 (1 v ) ( 1) 2 2nv 2 2nv n !(2 v )(3 v ) ( 1) 2 2nv 2 2nv ( n v ) (1 v ) n ( n v ) (2 v ) n n !(3 v ) ( 1) n n !(1 v )(2 v )(3 v ) ( 1) when r = v (n v) n n ! ( n v 1) ( n v ) (3 v ) 2 v (1 v ) 1 v 3 v (2 v ) 2 v 同理,當 r = −v c2 n set c 0 ( 1) 2 2 nv 396 1 v 2 (1 v ) n n ! ( n v 1) Two independent solutions of the Bessel’s equation 代入 When r = v cn x nr n0 Jv x n0 When r = −v J v x ( 1) x n ! 1 v n 2 n0 n ( 1) x n ! 1 v n 2 n 2nv 2 nv 稱作 Bessel functions of the first kind of order v and −v 6.4.1.3 Bessel function of the second kind 注意,兩個 roots 的差為 2v (1) 當 2v 不為整數時,Bessel’s equation 的解即為 c1Jv(x) + c2J−v(x) (也可表示成 c1Jv(x) + c2Yv(x)) (2) 當 2v 為整數,但 v = m + 1/2 (m 是一個整數) 時,Bessel’s equation 的解亦為 c1Jv(x) + c2J−v(x) (也可表示成 c1Jv(x) + c2Yv(x)) (3) 當 2v 為整數,且 v 是一個整數時, Bessel’s equation 的解為 c1Jv(x) + c2Yv(x) Yv(x): Bessel function of the second kind of order v (見後頁) 397 Yv(x): Bessel function of the second kind of order v Yv x co s v J v x J v x sin v 當 m 為整數時, Ym(x) 定義成 Y m x lim v m co s v J v x J v x sin v 用 L’Hopital’s rule 來算 sin v J v x cos v J v x J v x v v Y m x lim v m cos v 398 6.4.1.4 Bessel function of the 1st kind (order m 為整數時)的性質 (1) J0(0) = 1, Jm(0) = 0 for m 0 (2) Zero crossing 的位置,隨著 m 增加而越來越遠 (見 Table 6.4.1) 399 400 (3) J m x ( 1) J m x when m is an integer (4) J m x ( 1) J m x when m is an integer (5) d xv J ( x) x v J ( x) v v 1 dx (6) d xv J (x) xv J (x) v v 1 dx m m 見 Example 5, text page 263 6.4.1.5 Bessel function of the 2nd kind (order m 為整數時)的性質 (1) lim Ym x x 0 (2) Zero crossing 的位置,隨著 m 增加而越來越遠 401 402 6.4.1.6 Bessel’s equation 的變型 解:c1Jv(x) + c2Yv(x) x y xy ( x v ) y 0 2 2 2 2 2 2 2 (A) x y xy ( x v ) y 0 解:c1Jv( x) + c2Yv( x) Proof: Set t = x dy dy dy dt dx dx dt dt 2 d y Similarly, dx 2 2 dy dy 2 d y dt d d 2 dx dt dx dt dt dt 2 2 2 dy 2 d y 2 t 2 t t ( v )y 原式 = x y xy ( x v ) y 2 2 2 dt dt 2 2 2 t 2 d y dt 2 t 2 2 dy 2 2 (t v ) y 0 dt 對 t 而言是 Bessel equation y = c1Jv(t) + c2Yv(t) = c1Jv( x) + c2Yv( x) 403 (B) modified Bessel equation of order v x y xy ( x v ) y 0 2 2 2 解:c1Iv( x) + c2Kv(x) v 其中 I v x i J v ix 稱作是 modified Bessel function of the first kind of order v I v x I v x 稱作是 modified Bessel function of the Kv x sin v 2 second kind of order v 當 v 為整數時,也是取 limit 404 (C) x 2 y (1 2 a ) xy ( b 2 c 2 x 2 c a 2 p 2 c 2 ) y 0 解: a c c y x c1 J p ( bx ) c 2Y p ( bx ) 式子有點複雜,但可以用來解許多物理上的問題 Example 3 (text page 261) xy 3 y 9 y 0 x y 3 xy 9 xy 0 2 6.4.1.7 Spherical Bessel Functions Jv(x) 當 v 1 , 3 , 5 , 2 J1/ 2 ( x) J 1 / 2 ( x ) 2 2 sin x x 2 cos x x 2 405 時,稱作為 spherical Bessel functions 406 6.4.2 Legendre’s Equation 6.4.2.1 Legendre’s Equation (1 x ) y 2 xy n ( n 1) y 0 2 yx ck x k 代入,得出 (過程見text pages 265, 266) k 0 二個 linearly independent 的解分別為 n ( n 1) 2 ( n 2) n ( n 1)( n 3) 4 y1 x c 0 1 x x 2! 4! ( n 4)( n 2) n ( n 1)( n 3)( n 5) 6 x 6! ( n 1)( n 2) 3 ( n 3)( n 1)( n 2)( n 4) 5 y 2 x c0 x x x 3! 5! ( n 5)( n 3)( n 1)( n 2)( n 4)( n 6) 7 x 7! (a) When n is not an integer, both the two solutions have infinite number of terms. (b) When n is an even integer, y1(x) has finite number of terms. In y1(x), the coefficient of xk is zero when k > n. (c) When n is an odd integer, y2(x) has finite number of terms. In y2(x), the coefficient of xk is zero when k > n. y1(x) when n is an even integer and y2(x) when n is an odd integer are called the Legendre polynomials (denoted by Pn(x)). 407 408 通常選 c0 1 n/2 13 24 ( n 1) n c1 1 ( n 1) / 2 13 24 n ( n 1) (讓 Pn(1) 一律等於 1) 由 y1(x) P0 x 1 2 P2 x 1 3 x 1 2 4 P4 x 1 35 x 30 x 3 8 由 y2(x) P1 x x 3 P3 x 1 5 x 3 x 2 5 3 P5 x 1 63 x 70 x 15 x 8 409 Legendre polynomials Interval: x [−1, 1] 6.4.2.2 Properties of Legendre Polynomials (1) Pn x 1 Pn x n (2) Pn 1 1 even / odd symmetry Pn 1 ( 1) (3) Pn 0 0 when n is odd (4) Pn 0 0 when n is even n (5) ( n 1) Pn 1 x (2 n 1) xPn x nPn 1 x 0 recursive relation n 2 d 1 Pn x n x 1 n 2 n ! dx n (6) Rodrigues’ formula 410 411 (7) 1 1 Pm x Pn x dx 0 If m n orthogonality property (8) 若任何在 x [−1, 1] 區間為 continuous 的函式 f(x) 皆可表示為 f x a n Pn x n0 由於 f x P x dx a P x P x dx a P x P x dx m n n m n m m n0 根據 orthogonality property 所以 a n f x P x dx P x P x dx m m m Orthogonality property 才是 Legendre polynomials 最重要的性質 6.4.2.3 補充:其他常見的 orthogonal polynomial Chebychev polynomials 電子學和 filter design 常用 They are the solutions of (1 x ) Pm x xPm x n Pm x 0 2 2 1 1 1 1 x 2 Pm x Pn x dx 0 Pn cos C n cos n Hermite polynomials 電磁波、光學、頻譜分析常用 They are the solutions of Pm x xPm x nPm x 0 e x 2 Pm x Pn x dx 0 412 6.4.3 Section 6-4 需要注意的地方: (1) 概念簡單,但是定義,性質,和數學式甚多 擇要而學即可 (2) 要了解 Gamma function 413 414 Review of Chapter 6 解法適用範圍: Linear DE,且 coefficients 最好為 polynomials y (n) Pn 1 x y ( n 1) P1 x y P0 x y 0 當 Pm(x) 在 x = x0 時為 analytic x0 為 ordinary point yx cn x x0 n 代入求解 n0 當 Pm(x) 在 x = x0 時不為 analytic 但是 (x − x0 )n −mPm(x) 在 x = x0 時為 analytic x0 為 regular singular point 有時,另一個解為 yx C y1 x ln x n0 cn x x0 n0 bn x x 0 n r2 nr 代入求解 415 Exercise for practice Sec. 6-1: 2, 9, 14, 24, 28, 32, 34 Sec. 6-2: 2, 12, 13, 16, 20, 23, 26, 27 Sec. 6-3: 6, 9, 13, 18, 22, 24, 26, 29, 30, 31, 33, 36 Review 6 7, 10, 14, 19, 20