9* Matrices and Its Applications ELEMENTARY TRANSFORMATIONS (OR OPERATIONS} Any one of the following operations on 7 Ss ona matri {Gr E-operation). ix is called an elementary transformation (@ Interchange of two rows or two columns. The interchange of i'” and j' rows is denoted by Ry or R, oR, The interchange of i*® and jt® columns is denoted by C; or C0 6, (ii) Multiplication of (each element of) a row or column by a non-zero number k. The multiplication of i” row by k is denoted by kR, The multiplication of i" column by k is denoted by kC, (iii) Addition of k times the elements of a row (or column) to the corresponding elements of another row (or column), k #0. The addition of k times the jt» row to the ith row is denoted by R; + kR,- ‘The addition of k times the jt" column to the ith column is denoted by C; + RC, Ifa matrix B is obtained from a matrix A by one or more E-operations, then Bis s be equivalent to A. eee ‘Two equivalent matrices A and B are written as A~B. said to 2.2. ELEMENTARY MATRICES (M. J. Dee. 2012) ‘The matrix obtained from a unit matrix I by subjecting i tt to one of the E-operations is called an elementary matrix. 100 For example, let I=|0 1 0 001 100 (@ Operating R,, or C, on I, we get the same elementary matrix 00 1). 010 It is denoted by E,,, Thus, the E-matrix obtained by either of the operations Ry or Cy on Tis denoted by E,,. 100 (ii) Operating BR, or 5C, on I, we get the same elementary matrix 4 A 4 : *This chapter is not included in the syllabus of KU, Kurukshetra and DCRUST, Murthal. 73is abide Gir by 5E,, Thus, the E-matrix obtained by either of the operations R, Re TU (Gi) Operating R, + 4R, on I, we get the elementary matrix | 0 : " 0 Itis denoted by B,,(4). Thus, the E-matrix obtained by the operation R; + kR, is de by Bx) yy E,,(4). Thus, the “18 denoteq. a 100 (iv) Operating C, + 4C, on I, we get the elementary matrix 0 1 0},which is the trang. Oat 100 poseof|Q 1 4|=E,,(4) andis, therefore, denoted by E’,,(4). Thus, the E-matrix obtaineq by. 001 the operation C; + kC; is denoted by E’,(k). 2.3. THE FOLLOWING THEOREMS ON THE EFFECT OF E-OPERATIONS on MATRICES HOLD GOOD (@ Any E-row operation on the product of two matrices is equivalent to the same E-row operation on the pre-factor. If the E-row operation is denoted by R, then R(AB) = R(A).B. (® Any E-column operation on the product of two matrices is equivalent to the same E-column operation on the post-factor. If the E-column operation is denoted by C, then C(AB) = A . C(B). (©) Every E-row operation on a matrix is equivalent to pre-multiplication by the corresponding E-matrix. ‘Thus, the effect of E-row operation R,, on A = ‘The effect of E-row operation kR, on A = KE, ‘The effect of E-row operation R, + KR, on A= E(k) . A (@ Every E-column operation on a matrix is equivalent to post-multiplication by the corresponding E-matrix. Thus, the effect of E-column operation Con A=A. Ej; The effect of E-column operation kC, on A= A. (kE,) ‘The effect of E-column operation C; + kC, on A=A.. E’,(k) A ——_————————— OO OO 2.4. INVERSE OF A MATRIX BY E-OPERATIONS (GAUSS-JORDAN METHOD) The elementary row transformations which reduc a e 0 rn e a square matrix A to the unit mat when applied to the unit matrix, give the inverse matrix A~! Let A be a non-singular square matrix. Then A = IA Anely suitable E-row operations to A on the left hand side so that A is reduced tol a _ Bia sooot pot apply the same E-row operations to the pre-factor I on right hand sil 1=BA i B djot hee by elementary row operations. | 311 on DAS: 8 the inverse of A = | ng the given matrix side by side with unit matrix I, we get operating R, © R, operating R, > R,~3R, operating R, +R, ~2R,,R, >, +5R, 0 0 0 operating R, > R, 5-38 1 operating R, > R, +R, R, > R,—2R, oo as mo aswe get 3 a 2 3 aIf A is a non-singular n x n matrix, then p(A) [| A] #0is the largest minor of A.) IfT, is the n x n unit matrix, then | I, |=140 = pd,) =n. (iv) If Ais an m X n matrix, then p(A) < minimum of mandn. (@) If all minors of order r are equal to zero, then p(A) <7. ‘To determine the rank of a matrix A, we adopt the following different methods: (@ Start with the highest order minor (or minors) of A. Let their order be r. If any ong ‘of them is non-zero, then p(A) =r. Tf all of them are zero, start with minors of next lower order (r — 1) and so on till yoy get a non-zero minor. The order of that minor is the rank of A. This method usually involves a lot of computational work since we have to evaluate several determinants. (Gi) Convert the given matrix A into lower triangular matrix @ by elementary column operations or upper triangular matrix by ‘sg elementary row operations. ‘Then p(A) = number of non-zero columns in lower triangular der matrix or number of non-zero rows in upper triangular matrix. veo Anon-zero row or column has at least one non-zero element. ¢.. Sonam (iii) If A is an m x n matrix and by a series of elementary (row or column or both) operations, it can be put into one of the die following forms (called normal forms) : f eG) it, a -|,| |, : OJ, [1], where I, is the SRP 0: o}lo unit matrix of order r. Since the rank of a matrix is not changed as a result of elementary transformations, it follows that p(A) =r. [- rth order minor | I, | = 140] Note. For an m x n matrix A of rank r, to find square matrices P and Q of orders m and n respectively, such that PAQ is in the normal form le al Method. Write A = IAI. Reduce the matrix on L.H.S. to normal form by affecting elementary row and/or column transfor tie prefoctor on RS. : ele) Every eee ee transformation on A must be accompanied by the same transformation Example 3. Find the rank of the matrix ) __Buery elementary row transformation on A must be accompanied by the same transformation on ra res 2067. Gi)|1 4 2 . 2010) ed (M.D.U. Dee. te 1230 Qoct Rel(g'.4°9°/9 2-1 W)3 2 1 3 dine. 7. &A TEXTBOOK OF ENGINEERING MATHEMAriog eS He 0-3 -4 -8 0 0 23 -4 -8 0 =3 0 0 operating R, «> Ry eoracoens ‘The only fourth order minor is zero. Since the third order minor 1 0 “jo 0 1 0 ~jo 0 aerate 0-4 <8 60 -3 ranks ae, 2 tn. 3. 2 6 3 0-7 Sol. (@ Let =(1) 4) @3)=1240 -. p(A)=3. (M_D.U. Dec. 2009) A= Cone Senn GueEn Soon Soon Senn Example 4. Reduce each of the following matrices to normal form and hence, find their teers 6 Gj} 0 3 2 2) -8 -1 -3 4 3-1-1 ie # as operating R, > R, — (R, + R, +R) 38 0 -7 3-1 -1 wt 3 25| operating R, © R, Omror 10 -1 -2 -4 f = at operating C, > C, + C,, C, + C, + 2C,, Oa ©, C, +40, 00 0 i A A operating R, > R, ~ 2R,, R, > R, — 3R, 00 0 00 0 ‘ 9 0 operating R, > R, ~R, 00 0 0 0 0 1-6 -3 a 4 9 10| erating C, > C, + 6C,, C, > C, +3Ce oe!fi 2 i 12 i] nena nia Po as Oped Hence find the rank of A. (M.D.U., 2010, | 2] f1 0 0] f1 00 “2 g|-|0 1 o/Ajo 1 0 Pog i) [0 0 1) [0 o 2 3, — ©,, Cy -> Oy — 20, (subjecting the ‘post-factor on R.H.S. to same eoe’.-.2 100 A | Wee ajelo 1 ola To er 001 a 1 + R, -R, (subjecting the prefactor on R.HS. to same : Died) Ol [2-2 -2 jo aJ=|-1 1 0jAjo 1 0 Bera C, (subjecting the post-factor on R.HLS. to same operation) Peniefediso 0] [1 -1 -1 1 oj=|-1 1 ojAjo 1-1 coe |o° 0 1 | (subjecting the pre-factor on R.H.S. to same operation) aE 1 ola ao 1 =| 2011; K.U.K. 2013) operation)el 3 method to find the inverse of each of the following matrices: 20-1 5 wl fino) nee 1 01 3 5 2 -3 (K.UK. 2008, 2011, 2013) ite the inverse of the following matrices by using Gauss-Jordan method: 2 -6 -2 -9 + el aot 12 So ee iver3 1 - 411° % 0 1-39 I (K.UK. May 2012) ‘elementary transformations, find the inverse of each of the following matrices: 3 peloe es 125 3 G@i)\2 -3 4 (i)) 2321 o-11 -111 (G.B.T.U. 2010) ranks of the matrices 12 3 12 22 @|24 7 @ij|1 26 7 3.6 10 15 0 10 14 5 2-4 6 @|26 8 @i)|-1 2 -3 ‘ 3 7 22 3-6 9 j 3 -1 2 rank of the matrix | - 6 24 by reducing it in its normal form. -3 Le (M.D.U. Dec. 2014) et ae i 426-1 = Bris ank of the matrix 0)" '28 077 | by node ts miemnalinnems 16 4 12 15, the ranks of the following matrices: a 1-12 -3) 12,3 2 eer oro (i |2 3 5 1 of afi} «€ 13 4 5 o 10 2! (U.K.T.U. 2012) Sel65 a8 pes ee : w fen 4-8 wig? a 8 & 112 13 14 a gt? 16 17 18 19 41 23 -2 Se Oral woa-P = oe tae tak 2 matrices P and Q such that PAQ is in the normal form. Hence find the rank of A. (M.D.U. May 2008, Dec. 2012) oe 2 1-36 A=/3 -3 12 > unit matrix and verify that A“! = QP. Answers at 8-1-3 @)|-15 6 - @i|-5 1 2 5-2 2 10 -1 -4 ~21 0 1 Rit 0) 2-1 4 1-3 1 tet Oko a 2 pet er