Linear Differential Equations 13.1, DEFINITIONS: A linear differential equation is the differential equation in which the dependent variable and its derivatives occur only in the first degree and are not multiplied together. ‘Thus, the general linear differential equation of the n'” order is of the form a pe 8 . er Pa OE PE Py ok, where P,, P,, ... P,P, and X are functions of x only. A linear differential equation with constant co-efficients is of the form / pt 8 HS 59,7 tay 2 dx’ dx’ dx’ .,4,,)@, are constants and X is either a constant or a function of x only. ton Lr ayeX (1) where ay, dy, 13.2, THE OPERATOR D The part & of the symbol * may be regarded as an operator such that when it opor- ates on y, the result is the derivative of y. @ a Stalerly, may be regarded as operators, For brevity, we write & Thus, the symbol D is a differential operator or simply an operator. Written in symbolic form, equation (1) becomes ("4 aD + a,D™2 4 uu +a, D+ aby =X or ADy =X where f(D) =D" +a,D™ + aD’ ie. f(D)is a polynomial in D. The operator D can be treated as an algebraic quantity. Thus Du +v)=Du+ Dv, D(hu) =’ Du DeDtu= Du, — DPDtu = DADPu The polynomial f(D) can be factorised hy ordinary rules of algebra and the factors may be written in any order. tooo ta, D+a, 708LINEAR DIFFERENTIAL EQUATIONS 709 13.3, THEOREMS Theorem L. fy =) py =Jsyu1)'=),,are linearly independent solutions of the differential equation (Di +a D"! + aD! +... 4a, =0 ol) then w= ey, + Cy, ++ ¢,);, 18 also its solution, where ¢), ¢, ... ¢, are arbitrary constants. Proof. Since y = ¥,,.7 = Yano. =, are solution of equation (i). a Dy, +a,Dy, +a, D™4y, +... 44,9, =0 Dy, +a,D™y, +4,D"y, +... #4,% uli) vy, + aD ty, +a,D™%y, +. 40,9 Now D'u ta,D" tu + a,Diu et aye =D* (ey, + egy t+ 6,94) FAD Heyy, + yyy toot Cy3,) + QD Mey, + Cag Hot Cy) Fa ae oH MC, + Cag tot CY q) =¢D'y, +a,D™y, 4a,D"%y, +. 4 a,9,) +0,(D"y, +4,D"¥y, + a,D"%y, +...+ 4,9) to ot 6,(Dy, +a,Dly, + a.D'y, +..4 4,74) = o(0) + 6,10) +...4 6,0) [From @i)] =0 which shows that w = cy; + cq t....-+ ¢,9,,i8 also the solution of equation (i). Sinee this solution contains n arbitrary constants, it is the general or complete solution of equation (i). Theorem 2. ify = u is the complete solution of the equation ftD)y = 0 and y = v is a particular solution (containing no arbitrary constants) of the equation f(D)y =X, then the complete solution of the equation AD)y =X is y =u +o. Proof. Since y = 1 is the complete solution of the equation flDy = 0 w ADu sald) Also y =v is a particular solution of the equation (Diy = X (iii) AD =X iv) Adding (ii) and (iv), we have iD\(u + v) Thus y =u +v satisfies the equation (ii), hence it is the complete solution (C.S.) because it contains arbitrary constants. ‘The part y = wu is called the complementary function (C.F.) and the part y = v is called the particular integral (P.L) of the equation (iii). *. The complete solution of equation (iii), is|_y + Thus in order to solve the equation (iii), we first find the C.F. ie., the C.S. of equation (i) and then the P.1. ie., « particular solution of equation (iii). 13.4, AUXILIARY EQUATION (A.E.) Consider the differential equation (D" +a,D"1 4 a,D"2 4 ..... #a,)y = 0 A) Let y = e™* be a solution of (), then Dy = me™, D®y = me" ,....., DP 2y = mem710 ATEXTROOK OF ENGINEERING MATHEMATICS D'ty = mle", Dry = mre™ Substituting the values of'y, Dy, D4, ....., D'y in (i), we get Gm" saym™) sam? +... 4a,)e™=0 or mam! + ag”? + act @, = 0, since e™ #0 Ait) ‘Thus y =e” will be a solution of equation (i) if m satisfies equation (ii). Equation (ii) is called the auxiliary equation for the differential equation (i). Replacing m by D in (i), we get D" + a,D"" +4,D"~ +, (Git) Equation (ii) gives the same values of m as equation (iii) gives of D. In practice, we take equation (iii) as the auxiliary equation which is obtained by equating to zero the symbolic coefficient of'y in equation (i). Definition. The equation obtained by equating to zero the aymbolie co-efficient of y is called the auxiliary equation, briefly written as AE. 13.5. RULES FOR FINDING THE COMPLEMENTARY FUNCTION Consider the equation (D* +a,D'! +4,D"2+....+4,)y =0 “ where all the as are constant. Its auxiliary equation is D" +a," +a,D"2 +....+4,=0 wii) Let D= my, my Mg... m,, be the roots of the AE. The solution of equation (i) depends upon the nature of roots of the A‘E. The following cases arise : Case I. [fall the roots of the A.B. are real and distinct, then equation (ii) is equivalent to (D-m,) (D=my) .....(D=m,) =0 dit) Equation (iii) will be satisfied by the solutions of the equations (D-m,y = 0, D = myy = 0,.....5 (D-m,y = 0 d) Now, consider the equation (D— m,)y = 0, te. = =my=0 fmt me It is a linear equation and LP. lution is ye =[0.e"™ dere or y= ee™ Similarly, the solution of (D ~ m,)y = 0 is y = ce” the solution of (D — m,)y = Oisy = ¢,e!" Hence the complete solution of equation (i) is Y=eyo™ tego" 4... Hee" (iv) Case II. If two roots of the A.E. are equal, let m, = m, The solution obtained in equation (iv) becomes YEO +e) e™ eye +, cot +050 tne t Oyen It contains (n ~ 1) arbitrary constants and is, therefore, not the complete solution of equation (i).LINEAR DIFFERENTIAL EQUATIONS 711 ‘The part of the complete solution corresponding to the repeated root is the complete solution of (D—m,XD—my)y =0 Putting (D - m,)y =v, it becomes (D-m,)v =0 ie., As in case I, its solution is v = ee” (D—mpy which is a linear equation and LF. © itssolutionis ye" or y= (eye +5) o™ Thus, the complete solution of equation (i) is ise + ee my = my, then proceeding as y= (oye +e,) e™ + 050 If, however, three roots of the A.E, are equal, say m, above, the solution becomes Y= (ex? Hege + ey) oF ego Hoses t ge Case IIL If two roots of the A.E. are imaginary, let m,=a+i8 and m,=a-ip ‘The solution obtained in equation (iv) becomes foripe (iO 4 ego 4 ee pre + ene (eye + eg0P) + 60 + ane t Oye "[¢ (cos Bx + i sin Br) + ¢, (cos fix ~7 sin Bee)] $640 aes HOC = cos 8 +i sin 6 by Euler's Theorem, o = e™[(c,+ cy) cos fax + i (cy —c,) sin Px] + cye”** +......4¢,6" = e%(C, c0s fix + C, sin fix) + eye" +... ce taking c, +c, = C,, ile, 4) Case IV. If two pairs of imaginary roots be equal, let m,=m,=0+i8 and m,=m,=0-i8 ‘Then by case II, the complete solution is [(cya + ¢y) €08 Pav + (egx + ¢,) sin Bix] + ese™* ILLUSTRATIVE EXAMPLES 3 Example 1. Solve ae 2 7& ~6y =0. Sol. Given equation in symbolic form is (D* — 7D — 6)y = 02 A TEXTBOOK OF ENGINEERING MATHEMATICS Its AE.ism’—7m-6=0 or (n+1)(m+2)(m-3) whence m=-1,-2,3 Hence the C.S. is ce + ce + ce, Example 2. Solve (D'- 4D” + 4DJy = 0. Sol. The A.B. is m® — 4m* 4m=0 or m(m?—4m+4)=0 or mim —2?=0 whence m=0,2,2 Hence, the C.S is yee" ter +e) or y =e, +(e, +0,)e% dty d’y Example 3. Solve +1352 + 36y =0. pI arti gg t 88 Sol. Given equation in symbolic form is (D4 + 13D* + 36)y = 0 Tts AE. is m4 + 13m? + 36=0 or or m 2i, #3 % (es cos 2x + ¢, sin 2x) +e (e, cos 3x +c, Sin 3x) 4 €08 2x + cy sin Bx + es cos Bx +c, sin Bx. Hence the CS. is or y 4 Example 4. Solve G + 4x =0. it d Sol. Given equation in symbolic form is (D# + 4)x = 0, where D = —— dt lis AE. m'+4=0 or (m+ 4m2+4)— 4m? =0 or (m? + 2)%—(2m)2= 0 oF (mm? + 2m + 2m? — 2m +2) =0 cz - -22\-4 22-4 whence m= and “S35 ie, ma-1eiond 1s Hence the CS.is x= e"(e, cost +e, sint) +e! (c, cos t +0, sin). TEST YOUR KNOWLEDGE Solve the following differential equations dy dy 1 2 2 (a+) 2+ ady=0. Barer OTE aby =0 og @y ye 5. Sato Sp+uZ+oy=0 4. 2, 8 Fr +8£F + 167-0 9% (D+ 19D? +D+Dy=0. Px ade de 10, Sy -8 G+ Bea 0, given that when f= 0, x= 0 andLINEAR DIFFERENTIAL EQUATIONS 713 a ‘ Y 45 1, Ted + 205 =, given that whens = 0,y =O and 5 = 15, a@x 12, 1 £% = mtr, show thatx =e, cos me + ¢, sin mt +e, ar * Answers L beet cet beg edt NDE 4 eget 4. tepew et ep teeter eet tee tee T. yacet beet teehee ® Ry = le, +6,x) c08 Be 4 (ey + 6,9) sin Be ols 5 oy tops + eget cone + (ey toe tepMeine re ® [= +e) ens Fa + le +2) sin AL. y = Se sin 5y. 1 13.6. THE INVERSE OPERATOR 7 Definition. —1_ x is that function of x, free from arbitrary constants, which when operated upon by A(D) gives X. 1 1 x\. Thus 1 {ry 8} =X v AD) and a are inverse operators. 1 Theorem 1. 75 X is the particular integral of f(Diy =X. Proof. The given equation is (Dy = X A) Putting 1 {1 Xi z x} = ich is FD) eh wenave AD) Fe) } X or X=X, which is true. al X is a solution of (1). Since it contains no arbitrary constants, it is the particular integral of f(D) y = X. ‘Theorem 2. Prove that, $ X= fxdr. 1 Proof. Let pe Operating both sides by D, we have D ( Integrating both sides w.r.t. 2, fxae, y714 ATEXTBOOK OF ENGINEERING MATHEMATICS P : . 1 P no arbitrary constant being added since y = 75 X contains no arbitrary constant. $x=| xan Theorem 3. Prove that, =— D-a Proof. Let K= Da” D-a Operating on both sides by (D — a), (D —a) ( or x= Bea ie, Bg dz dx alk , feats which is a linear equation and LE. =e!" = e-@ Its ototionis yee = [ Xew de, no constant being added or y=e [Xewds 1 Hen’ lence: Dos Kee" foe Xde 13.7. RULES FOR FINDING THE PARTICULAR INTEGRAL Consider the differential equation, (D" +a,D"1 4 a,D"? + It can be written as f(D)y = Case I. When, Since Die cat (D" +a," +a, D"" +... +a, D+a,)e" =(a" +a, a"! +a, a ta, ata,)o% or AD) e* = fla) ew Operating on both sides by = ap For) 7" or eo =fla) 7 en Dividing both sides by fe), =*~ aa ie a *, provided fla) «0 a fla)” Hence provided f(a) #0.LINEAR DIFFERENTIAL EQUATIONS 715 Case of failure. If fla) = 0, the above method fails. Since fla) = 0, D = ais a root of AE. AD) = 0 D~aisa factor of AD). Let AD) = (Da) 0 (D), where § (a) #0 eli) ‘Then «_—_1_ #4, (D-a)o(D) D-a oD) D-a ota) 1 1 a, =e fe™ ods y say Daa” oar Je e [by Theorem 3] 1 gos _ Low - a0" frdeme oe wldi) Differentiating both sides of (i) w.r.t. D, we have f’ (D) = (D~a) ¢’ (D) +9 (D) = Fla) = 61a) From (ii), we have <1 provided f’ (a) +0 If provided f(a) +0 and so on. ILLUSTRATIVE EXAMPLES Example 1. Find the P.I. of (4D? + 4D — 3) y =e. 1 2x 1 2x . Sol. ha 2 I D by 2: " WD?y4D-3° ~ 4@?14Q)-3 Geplaaine nay 2) alge a Example 2. Find the P.J. of (D? + 3D + Qy = 5. Sol. PL= + — (56%) fy ea] DP+3D+2 5 : "3 (replacing D by 0) Example 8. Find the P.J. of (D° 3D? + diy =. L Sol. PL. = ——,—e™. D?-3p +4" Hence the denominator vanishes when D is replaced by 2. It is a ease of failure We multiply the numerator by x and differentiate the denominator w.rt. D. Lox PL=2.—)_ 3D? -6D716 ATEXTBOOK OF ENGINEERING MATHEMATICS It is again a case of failure. We multiply the numerator by x and differentiate the de- nominator w.r.t. D. Pi=x?,—1_ = .2,_1 @-6 62-6 Case IL When X = sin (ax + b) or cos (ax +b) Dsin (ax +6) =a cos (ax +b) - a) sin (ax +b) @ cos (ax +b) if sin (ax +5) in (ax + b) in (ax +B) Dé sin (ax +b) or (D?)? sin (ax + b) (- a®)? sin (ax + b) In general, (D)* sin (ax + b) = (-a?)' sin (ax + b) S AD*) ain (ax +b) = f (—a2) sin (ax +b) Operating on both sides by FD)" 1 (fD*) sin (ax +5) = Ap (fa?) ain (ax +B) sin (ax +) =

m=tia CF. = tan ax. €, cos ax + ¢, sin ax 1 Dried ia 1 PL.= peg tanax tan ax (Partial Fractions) il mane| int sins of taartesar—tsinon dene sinex 182°) a eee isc ote f [sin ax 1 22°82) dew lf [gin x — ioce ax cos a8] de ( vos = via [ SOS GE _ + tog (see ax + tan ax) +i 22% a 3 1 = elt [(cos ai sin ax) + # log (see ax + tan ax)] 1 . 1 ao" fe“ + i log (sec ax + tan ax)] = io [1 + te’ log (sec ax + tan ax)] 1 1 ” Changing i to -i, we have tan ax = —— [1 - ie“ Jog (see ax + tan ax) Dtia a a+ Bia a + log (see ax + tan ax) ( a 14+ ie! log (sce ax + tan ax)} ++ {1 ie“ log (accax + tan |} a log (see ax + tan ax) . cos ax Hones the C.S. is y = ¢, cos ax + ¢, sin ax —LINEAR DIFFERENTIAL EQUATIONS TEST YOUR KNOWLEDGE Solve the following differe mntial equations : ay a Gets see ay & o5 a Sh 4 8 45) --2eoshe @y @y dy BTS Tat Eye sin de 7. (DP 4D + B)y = sim Sx cos 2x 2. % Baagysersinge 4. 13. 15. (DE 8D + 2y = 26 e085 i. 19, (D?4 4D + 8y = 120% sin sin Bx 21, (D¥ 42D? + Dy =2% + sin? F x 28. (D-DAD+ Diy =sint 5 +e +e 25. (D2 Ly=xsine + (4x40 27, S24 4y=4tan oe ds! z 6, —X + y= sin Bx eos? = ae 2 (D* -SD + 2yy = Ge + sin 2x Dg oy =e 4 sn 20 ds 18. (D? — 2Dy = o* sin a ‘ 20. be 4 Dy = x48" 4 €¥ c08 Dx 22, (D® —4D + 4)y = 8x? e% sin 2x 725726 ATEXTBOOK OF ENGINEERING MATHEMATICS 5 6. 10. 15. 16. i. 19. 20, 21. 22, 23, 2, 25. 26. 27. 28. 1 ¥ YEQeT + eyes sey sin x + Fe (2-c0s Ze— sin 2x) 1 * 8 is 1 402 | cy cos +esin + oo (sin Bx + 27 €08 3x) — este” | ep cos 9 x bey sin 5x | +25 (sin Bx +27 eos 3x) 1 334 20 3 10 1 y= eye + eye + Fr (0 cos Bx ~ 11 gin Gx) + 5 sin x +2 conx) 1 ya eyet + 6% + Ty CH + GG (B 008 Bx — sin Be) 1 J =e; 008 2x + ey sin Bx + Fe —F con Be y =e, +e (cy cos VBx +c, sin V3 x) +2 + sin 20) IB y =e, + ge + ee + Peon ae 26x yy moet + 046% yaeet+ce™ + inv . VEC +ep* +0, cos e+e, sinx—= e* cos.x IB. y=, teye™— > e sine 5 28 (c, cos Be + 6p sin 2x) + 4 e-® (Be sin Qe + cos Ax) 2 as x cos axe (2-22 2).e 2 Besc0g. 00s 2x +c, 8in V2x+ o(¥ art yar) a7 (sim 2e— cos 2x) 2x 2,2 7 11 1 ae, 4 (cy reget + 4 ( 22-24 DB) sin 20+ 4.008 20) yesetetepers S42 (tLe dl) od tein nertenc tl y= (ey Heya e™ ctx cos 2x + (2x? 8) sin 2x} 2 x cosas etx 1a Yale, Heme’ +c, Hee + Ty 1 y me, 008 Qe +e, sin Bx + > (Bx sin x — 2 cos.x) 9 L 1 ym eyet + eget le sin x + cos x) + zy set Qs? Be +9) log cos ax T*) cos 2x + ¢, sin 2x ~ cos 2x log (see 2c + tan 2x) i Vf y =e, cosax-+e, sin ax + —{xsin ax + evs ax y 2 tem ot.LINEAR DIFFERENTIAL EQUATIONS 727 13.8. METHOD OF VARIATION OF PARAMETERS TO FIND P.l. Consider the linear equation of second order with constant co-efficients ay, FY a ® say =X eth Gg tw Let its C.F. bey = ¢,y, + ¢yy, 0 that y, andy, satisfy the equation dy dy Fr 4,2 +ay=0 dx 1 ae gy Now, let us assume that the PI. of (1) is y = uy, + uy. where u and v are unknown functions of x, Differentiating (3) w.r-t.x,wehave assuming that w, v satisfy the equation u’y, + vy, 0 Differentiating (4) w.r-t. x, we have = uy," uy! + vy94 0" Substituting the values of y, y’ and y” in (1), we get (uy "+ w'yy! + vy” + Uys!) + a,(uyy’ + vY,/) + agluy, + Uy,) or Uy," + O94 + ayy, + UY,“ +a,’ +ay),) + Uy +07 or since y, and y, satisfy (2), Solving (5) and (6), we get wa|y 22, yo! ya [% x| ls and v= . 1. YS ly x|*| yy where W=|7, 22,] is called the Wronskian of y,, 75. NM 2 : _ fe yiX Integrating ee ee Substituting in (3), the P.I. is known, iy + v9! +U'y, + Uy = yy! tye! ~Q) onl) (3) wl) (5) (6) Note. As the solution is obtained by varying the arbitrary constants ¢,,; of the C.F., the method is known as variation of parameters. 2 Example. Apply the method of variation of parameters to solve ay +4y =4 sec? 2x, Sol. Given equation in symbolic form is (D? + Aly = 4 soc? Qe Its AB. is m?+4=0s0that m=+2i os ORis y= €08 2x + ¢, sin Be Here y, = cos 2x,y,=sin 2x and X=4 sec! 2x cos 2x sin 2x -2sin 2x 2eos 2x728 ATEXTBOOK OF ENGINEERING MATHEMATICS == 20s 2x f sec 2c tan 2edx + 2sin 2x f see 2x de == 20s 2x , 580% 1 +2sin 2x. 5 log (see 2x + tan 2x) =— 1+ sin 2x log (see 2x + tan 2x) Hence the C.8. is y = ¢, cos 2x +c, sin 2x ~ 1+ sin 2x log (sec 2x + tan 2x). TEST YOUR KNOWLEDGE Solve by the method of variation of parameters : dy 1 22 4y-cosees. 2 ac? *” 8 he ayo tan 2x 4 Gs pessins 5. y'-2y' #2yse tame Answers 1 Ske, Sov Roe eR 2 y=c,cosx+0, sin x c08 2 log (cos) +. sin x By =e, 608 2e-+6, sin 2r— + cos 2 log (ser 2x + tan 20) i * . doen +0 dines Sein" cose 4 ssines 3 5. (e,€08:« +6, sin x) ~ 6 cos. log (soe x + tan 13.9. CAUCHY’S HOMOGENEOUS LINEAR EQUATION An equation of the form x a +a,xtt S sagt tae & 4ay-X where a,'s are constants and X is a function of x, is called Cauchy's homogeneous linear equation. Such equations can be reduced to linear differential equations with constant co-efficients by the substitution x=e or z=loge so that =| orLINEAR DIFFERENTIAL EQUATIONS 729 3 Similarly, x° © = pep ~ 1 — dy and so on. de Substituting these values in equation (i), we get a linear differential equation with constant co-efficients, which can be solved by the methods already discussed. ILLUSTRATIVE EXAMPLES Example 1. Solve x? £ % Sol. Given equation is a Cauchy’s homogencous linear equation. Put x=e ie, log x dy ad’y so that x— =Dy, 2? —— =DOD-I1y de ae » dy d fm ae — Dy, = #8 CF = DD~ 1D ~ 2p, whore D Substituting these values in the given equation, it reduces to (D(D - 1)(D - 2) + 2D(D - 1) + Dy = 10? + e*) or (D? — D2 4 Dy = 10(e +e) which is a linear equation with constant co-efficients. Its AE. is m?—m2+2=0 or (m+1Xm2—2m+2)=0 24Ja-8 3 Llsi «ant ‘ CP. =c,e* +e (c, cos z +c, sin z) = + + xe, cos (log x) + ¢, sin (log x) rere ats 24 ‘| PI.=10 D-p+2° “DP -p?v2 D°-D?42 1 1 1 1 =10| +e? +2. — ee |= 10 2 +z. + _ 20 (sas **-3p? ap © } (¢« +230 CD ) 2 = Bet 4 De® = Bx + = log 2 “1 2 Hence the C.8. is y=“! + xle, cos (log.x) + ¢, sin (log x)] + 5x + — log x. ¥ 2 Example 2. Solve 2° s = Sol. Given equation is a Cauchy's homogeneous linear equation. - By =x log x. . dy 2dy af Put x=e Le, = log x so that 25. qe Dy, whore Da730, ATEXTBOOK OF ENGINEERING MATHEMAT! Substituting these values in the given equation, it reduces to (DID -1)—D—3]y =z or (D*- 2D ~3)y = 20% which is a linear equation with constant co-efficients. Its AE. ism®-2m-3=0 or (m-3)m+1)=0 m=3,-1 CF. = ce + eg¢7 =¢,x3 + = 1 2s 26 1 aot Pl. Sree D*+2D-3 = D?-ep-3 ° (D+2))-2D+2)-3° Hes]: 13.10, LEGENDRE’S LINEAR EQUATION An equation of the form 1 mt a + ber FY sales bet Pd sa, tbs) % ra,y-X ra iP ax where a/s are constants and X is a function of x, is called Legendre’s linear equation. Such equations can be reduced to linear differential equations with constant co-efficien : . dy _dy dz__b_dy by the substitution a +bx=e" ie, 2=log(a+bx) so that Se = 9°. So =o” a =b Dy, where D=7~ or Body, 6 déy dy (a+ bx)? dz a+bx’ dz” dx a, by aly wy ~ (a+bx)? dz atby dz? atbr (atbr\d2® dz d’y 4, 4 or (a+ bay TF = 02 (Diy — Dy) = 6 DID — Yy le 3 Similarly, (a + bx) a Bs DD-1M-2y. Substituting these values in equation (Z), we get a linear differential equation w' constant co-efficients, which can be solved by the methods already discussed.LINEAR DIFFERENTIAL EQUATIONS 731 Example. Solve (32+2? 22 43(3x+2) © —a6y =a22 +4n41 dx? de Sol. Given equation is a Legendre’s linear equation. Put Bx+2=e ie, z log (Gv +2) sothat (3x +2) 2 d (r+ 2° £2 =9°DD-1y, whereD= 4, dx az Substituting these values in the given equation, it reduces to eDD—1~2ap—a6y =2 3 Lied or 9D? — Ay = se =a or (D? - dy Jie SOT which is a linear equation with constant co-efficients. ItsAE.is m?-4=0 . m=a22 OF. = ee + ep = 0, (Bx + 2? +6, (Be +272 -t 1 Gx PL= oF DF otl[2e+t]-teet eye 2 = [fe +t)-he + D= oy [Gx + 2) log (ar +2) +11 : oi Hence the C.S. is y =¢,(3x + 2)? + ¢,(3x + 2)? + Jog (3+ 2) log (3x + 2) + 11, TEST YOUR KNOWLEDGE Soive: by aoe 5 be wg Lx? Ga t Oe + By = 50 2, 3 2 Dae ao 2cc+ 08 2, 6 Theradial displacement 1 in a rotating dise at a distance r from the axis is given by? S +r ft dr —u + br = 0, where b is a constant. Solve the equation under the conditions u = 0 when r = 0, u=0when r=a.732 A TEXTBOOK OF ENGINEERING MATHEMATICS 2. 2 EE soe eB day =x loge ro. fae yest 2tog tix Fea ® say sin (log x) wets at Oh Beye 65 cos (log x) 13. of ax 4 5y <2? sin Cogn 15. eres 16. (1 +2? HE 14) ® +y=Aeos logit +2) In +208 > ay = 60429 Bs 1ey= 81+ 2 * dy pty & _ i= 18. (2x43)? SF ~ 12x +8) 5 ~ 12y = Ge, Answers Lyre, cos @ loge) +ezsin Blogv1+2 2y eon 2 & yeorteegs- = 2 1 Fogel + 49 20 a By =le, +e log ade teed + = loge ii) y= 0,82 + ox — x2 log 2a ny (c, +0 logzhe + log x +2 ff [ey eos (og x) +6, sin Qog2)] +x logs 9. y=ea3+eqrt+ Flog x (7 log.s—2) 2 10. y =eyrt + oct 7 1 11. y =2%e, cos (log 2) +c, sin (og 2] + [sin (log x) + cos (log x)] 12, y =eye® + ele, cos (VB log x) +e, sin (YF log.2)} + 8 cos (log x) — sin (log x) 1 13, _y =.a%le cos (log 2) +e, sin (log 2)] ~ 3 a log x cos (log =) 2 Ie, cosh (V3 log x) + eg sinh (vB log x)] L % 2 1 ‘i [lo 215 sin (log x) + 6 cos (log x) + (27 sin (log 2) + 191 c0s (log | te 15. y =e loge +c, +2 logx)? 16. y =c, c05 flog (1 + x)) +c, sin [log (1 +.2)} + 2 log (1 +x) + sin log (+2) 17. y =(14 2v Plo, + 0, log (+ 2x) + flog (1 + 239/41 18. (2x +B) 1 4 Qe 4 3)3— 9 (Ox + 8)43,LINEAR DIFFERENTIAL EQUATIONS 733 13.11, SIMULTANEOUS LINEAR EQUATIONS WITH CONSTANT CO-EFFICIENTS Now we discuss differential equations in which there is one independent variable and two or more than two dependent variables. Such equations are called simultaneous linear equations. To solve such equations completely, we must have as many simultaneous equations as the number of dependent variables. Here, we shall consider simultaneous linear equations with constant co-efficients only. Let x, y be the two dependent variables and ¢ the independent variable. Consider the simultaneous equations fDw+fDy=T, 1) and 0,(Dx + oD =T, 2) where D= z and T,, , are funetions of f. To eliminate y, operating on both sides of (1) by @,(D) and on both sides of (2) by /,(D) and subtracting, we get [A(D) 0,(D) - 6D) /D)1 x = ©D)T, - {D)T, or AD) x=T which is a linear equation in x and ¢ and ean be solved by the methods already discussed. Substituting the value of x in either (1) or (2), we get the value of y. Note. We can also eliminate x to get a linear equation in y and ¢, ILLUSTRATIVE EXAMPLES. dx dy ig ® 19 Example 1. Solve 7 + 4s-48y =, 7) + 2x + By Sol. Writing D for £ , the given equations become (D + 4ix + By = AL) and Qe + (D+ By 2) To eliminate y, operating on both sides of (1) by (D + 5) and on both sides of (2) by 3. and subtracting, we get ((D + 4D +5) — 6x = (D + 5-30 or (D? + 9D + 14)x = 1 + 5¢- Bet Its A.B. is m2 49m +14 or (m+2Xm+7)=0 + m= CF. = eye + ce i 5i—3e! PL = peyapqaa (1+ 51-8) =— tt — 45,1 _+-3_, 1 Di+9D+14 D+9D+id) "DP +9D+14 1 1 1 = 5, 1-8 F +9014 © 9D, D® Prope © q4fie + "aa "1a 1 2 i. 9D | D? Ls E,& 9D D' atte Sef a4 f1-[= ls wa ta| (44 14 Bo 14! 14 “14734 ATEXTBOOK OF ENGINEERING MATHEMATICS Now dx Substituting the values of x and “j- in (1), we have an Bt 2ee + Tee %—- +e 1 ye Fk 2eye* + 8e,e7 5 Hence waco +eel+ —t- , * 4 2 1,,9,5 HS cettee tits ea verge a 98” 2: 2 2 o dy de Example 2. Solve TF + defy = 4Bx + dy =t41 ie Sol. Writing D for 4. the given equations become (D? + 4)x + 5y and Be +(D?+4yy 0+ (2) To eliminate y, operating on both sides of (1) by (D? +4) and on both sides of (2) by 5 and subtracting, we get A) [D2 + 4)? — 25] = (D? + 42 — Bit + 1) or (Dé + 8D? — 9x = 2 + de? Gt -5 = 40? 5t- 38 Its AE. is m4 +8m2—-9=0 or (m? 4 9m?—1)=0 mal e8i CF. = ce! + e,e7 +c, cos 3t +e, sin 3t 1 2 = Dr eprog 8-8) = at t=) fy | 1 sb? Dt a (4? — 5¢-3) 2 pf a4 {SP DY, | ator) 9 9° 9LINEAR DIFFERENTIAL EQUATIONS 735 (40? -or 4 ef + ye + 0, c08 3t +e, sin Bt 5 =-i 40 -01-3+8.6)|- 8,5 Now S weet *—3¢, sin 34+ Be, cos BH Gt +5 7 8 TE x oe! + 0x0! — 90, cos 3¢—9e, sin 3t— > dt? . 9 ax Substituting the values of x and SF in (), we have t dx db = t?— dee! eye — de, cos 3t — de, sin 3t + 16220), MB ot cot + Sey cos 31 + 9e,sin 3+ © 9° 9" 81 9 ye i+ Beye! — Beye + Bey cos St + 5c, sin B+ Be Bee +2 5 81 Hence x= cye + cet +c, cos St +c, sin Bi — 3 {ae -5t+ 3) Ad ys-ee—ce "eens Bt eg sin ae + 2 (ae ~ Hae 4) Example 3. Solve the simultaneous equations a ta tye Ot 2 +x =O given x(1) = 1, 9(- 1) =0. , dz Sol. The given equations are tap tys0 wf) dy 3 toh +e=0 (2) Differentiating (1) w.r.t. ¢, we have 2 pede, de dt’ dt Multiplying throughout by t o@x dx dy dt? tf dt te dt =o 2 or Poe, ,@ yoo [Using (2) ...(3) ae ae which is Cauchy’s homogeneous linear equation. Puttingt =e ie, w=logt, sothat 2-4 - rp, equation (8) becomes dda [D(D-1)4+D-1]x=0 or (D2-1)x=0736 ATEXTBOOK OF ENGINEERING MATHEMATICS Its A.B. is m?— 0, whence m=+1 e xecyel teeta ott wld) dx ee =-+G--+ ete (5) From (1), yootae (« ott (6) Since x(1)=1, from (4), wehave 1=c, +e, Also y(- 1 from (5), wo have 0 =¢,~cy Solving Hence Example 4. Solve the following simultaneous equations de dy de a ae ae Sol. The given equations are dx ry @) Differentiating (1) w.r-t. ¢, (Using (2)} pegs OS G8 asa Differentiating again wat.t, 5 =407 = dis) it or (D5 ~ 8x =0, where Ds Its AE. is 8=0 or (m-2)(m?4+2m4+4)=0 whence m=2,-12i V3 woeereapren Garg 1 nam, godt 2 dt if [Beye — eq 6° 008 (Bt 09) ~ egBe sin (BH —e9) 2H + eget [eos 3 cos (/3t ~c,) -sin an Lay From (2), ar 1 2 -t Qn . Qn = g]rere™ —en0 cos | VB — en += = ¢ V3 et sin Vt egLINEAR DIFFERENTIAL EQUATIONS 737 Qn an 24, | cos cos| Bt ~c, +72) se 6p eos co io, : a 14 eget cos (Bre +2). TEST YOUR KNOWLEDGE Solve the following simultaneous equations : is 2 6. dy 3 1, Gs 20-49 = 0; given that x == 0 when f= 0 4A, (D+ s+ QD + Dy =e, (D— De + (D+ Dy as dy 8 Bis res ay 0,904 D 4 ay = 202. ate ae 6. (D~ Lx + Dy =2¢ +1, (2D + Lx + 2Dy dx a 9. ‘y 10. SE syasint, £2 ea -cost ae 11, A mechanical system with two degrees of freedom satisfies the equations ax dy oe gh de” dt i 5 dx dy Obtain expressions for x and y in terms of t, given x, y, 2, & ati vanish att = 0. p y ei 9 oe adr = Answers 1, x=eM(c, cos t + eg sin), y =e lle, — cg) €08 6 + (c, + ¢4) sine] raosetyaet—el4sint 2, 1 1 2 x=-57 1+@e%+ = 143 Ht. (2-5 ay (1+ 60) ay +99, +80 eM +5 2-30) 2 1 1 1 Powherea= 5 G+ ViT).b= 3 BVT) 1 = 5 bef 1 acye"" + bee. y = eye" + ce wag le! 1-acye" + bee y = ey 1,8 48 7 6 oop h—F My ee — L Raley tebe’ te reWey=—F ley rey (L 4 Ole #5 ley egle! 2 -3: 1. -% ac, +e:bel + eye 2 —t, y= legB—1)—cjle— F eye 2738 ATEXTBOOK OF ENGINEERING MATHEMATICS 1 yet ye Te (cost +3 sin) 3 3 9. x= See! —seye% + ce! + S ot (eost-2 sin, 9. x= Sep! See + oye + TF of (cost —2sin 0,9 t 10, x=! + cy + ¢5 eos Cre sin f+ 5 (Gin ens) A 1 ‘ ya oye eget + cy cost + oy sint + 7 +6) (sine - cost), i. 13.12. TOTAL DIFFERENTIAL EQUATIONS An ordinary differential equation of the form Pdx + Qdy + Rdz = 0. AD) where P, Q, Rare functions of x,y, is called the total differential equation. ‘The necessary and sufficient condition for integrability of (1) is aQ_ aR aR aP aP_9Q (2-3) +o EF) + +6(F-2)- e 13.13, METHOD FOR SOLVING Pdx + Qdy + Rdz=0 Method I. Solution by inspection. When the condition of integrability is satisfied, it may be possible that by rearranging the terms, it becomes exact and the solution is found readily. ILLUSTRATIVE EXAMPLES Example 1. Solve : (yz + 2x) dx + (2x — 22) dy + (xy -2y) d2 = 0. Sol. Here = yz + 2x, Q= 20-22, R=xy- {2-2)42-2-2) Ox Oe dy dx = (ye + 20) (v2 x + 2) 4 (ex ~ 22) (y —y) + (xy — 2y) (@-2) = 0 ‘Thus the condition of integrability is satisfied. Regrouping the terms, we can write the given equation as (yadx + zxdy + xydz) + 2xdx — 2edy + ydz)=0 Integrating, we got xyz +.x%— 2yz = ¢ which is the required solution Example 2. Solve (x + 2? dy +y? (dx +d2)=0. Sol. The given equation can be written as ydx + (x +2)" dy + y%dz =0 Here P=92,Q= +22, R=y? 29_aR) (28 28), (28-2) } dy Ox = yP[2tx + 2) - 2yl + (x + 2)? (O- 0) + y*2y - 26 +2)] = yx + Bz — By + 2y—Qv—22)=0 ‘Thus the condition of integrability is satisfied.LINEAR DIFFERENTIAL EQUATIONS 739 ‘The given equation can be written as oy 4 aes 0 yl (ete xiyte 1 or - = y xtz yarta)¢ or clx + y +2) + ylx +2) = 0 which is the required solution Method IL One variable regarded as constant. In case the condition of intograbil- ity is satisfied and any two terms, say Pdx + Qdy = 0 can be easily solved, then the third variable z is taken as constant, so that dz = 0. Solve the equation Pdx + Qdy = 0 by any of the methods discussed in chapter 11. Let the solution be u = c. Replace the arbitrary constant c by any arbitrary function of z, say o(2). Thus w = (2) is the solution of Pdx + Qdy = 0, where o(2) is a function of z only. Differentiate u = o(2) w.r.t. x, y, z and compare the result with the given equation to determine (2). Example 3. Solve y2 dx ~ 2x2 dy + (xy ~2y?) dz = 0. Sol. Here P=yz, Q=—2xz, R= xy- zy" aQ aR aR aP (aP aq (2-3)-E-2)5- 2) = ye(~ 2x —a ~ 8y%2) ~ 2ealy ~ 9) + (ay ~ 2y*Me + 22) 2(— Bx + By%2) + Belay —zy") = 0. ‘Thus the condition of integrability is satisfied. Considering z as constant so that dz = 0, the given equation becomes yadee — 2xzdy = 0 Integrating, - ds gd 9 x oy Integrating log x—2 log y = log ¢ a 2 or ee 3 eHey yx Replacing ¢ by (2), we have x=y29(2) or x~y? G(e)=0 oD) Differentiating (1) w.r-t. 2, y, 2, we get 1dx - 2y0(z) dy—y?o'(z) dz = 0 wf) Comparing (2) with the given equation, we get = e paring given eq Bel 2902) yD) 2 x-2y or see -B) From the first two members, we have x = y6(z), which is the same as (1) ‘Thus, we don’t get any information about $12). ‘Taking x we have Ay =F yes [Using (1)] ye z740 ATEXTBOOK OF ENGINEERING MATHEMATICS de 1 ise ee = 1 which is a linear equation. » ge ta® & o@)=federe=F +e or o= 542 or or 2ny =" which is the required solution. Example 4. Solve 2yzdv + zxdy — xy (1 +2) dz = 0. 22+ 2c) or Baz =y%2? +O), taking 2c = C Sol. Here P=2yz, Q= 2x, R=-xy(1+2z) 2Q_ aR), 9(/2R ar), _/aP_2@ is ce el we) B =| = 2yz [x +x + x2] + 2x [-y— ye 2yl —ay(1 +222 -2) = Aaya + Dayz? — Bxyz — xyz? — ay2 - a2? = 0 ‘Thus the condition of integrabili isfied. Considering z as constant so that dz = 0, the given equation becomes 2yzdx + zxdy = 0 dz dy or 2 a0 x oy Integrating 2logx + logy=loge or zy Replacing ¢ by o(z), we have x2y = 9(2) “D) Differentiating (1) w.r-t.x, y, z we get 2xydx + x2dy - o(@)dz = 0 (2) : : 2yz _ 2x _ xy(1t2) Comparing (2) with the given equation, we get >= === = 2072 paring given eq ato sO) 2 é 2.904) 9 ye Eyre x #@ z or a= (2 + 1) [Using (0)] z or ¥@_1,) o@) 2 Integrating, log lz) = log z + 2 + loge = log ce or (2) = eze or aly sexe? Method III. Method of homogeneous equations. In case P,Q, R are homogeneous funetions of x, y, 2, then one variable, say 2, may be separated from the other two by the substitution x = uz, y = vz as illustrated by the following examples. Example 5. Solve (y? + y2) dix + (2? +2x) dy + (7 — xy) dz = 0. Sol. Here P=y? +92, Q=2% 42x, R=y®-ay (aQ aR oR oP aP_ aQ See ae a) = (2 +2) Qe + ~ By +x) + (2? + zal-y —y) + (y" —ay My + 2-2) Ay? + yall —y +2) — Ile? + 2x) + By? xy) = 2ay?—y? + 9% + xyz — yz +92? ye? — ay2 +9 ~a9' =0LINEAR DIFFERENTIAL EQUATIONS 744 ‘Thus the condition of integrability is satisfied. Since, P, Q, R are homogeneous functions of x, so that de=zdutudz and d; ‘The given equation becomes z, we substitute x = uz andy = vz dy + dz 2%(u? + vzdu + udz) + 271 + uedu + vdz) + 2%v* sw)dz = 0 or a2l(v? + oldu + (1 + wide] + [u(v? + v) + o(1 + u) + 0? -wldz = 0 or 2zivv + Ddu + (1 +u) do) + (uv? + uv + 0 + v?Jdz =0 or zlv(v + lidu + (1 + u)du] + (vu? + vu + dz = 0 1 1 or i oe Et le utl vu+) 2 ut+1 lo v+l z Integrating, we have log (u + 1) + log v — log v +1) + logz = loge (w+ Doz or Gree vt] or (Ess}y-e(2 3) z 2 or (+ zy sey +2) TEST YOUR KNOWLEDGE Solve the following equations : 1 rdetzdy+(y+22d2=0 2. yz dx tzxdy +xy dz =0 BG te)dr+le+addy+ (x ty)de=0 4, (yz +.aye) de + (2x + ays) dy + lay + xy2) d: 5. yz logz dx—2x logz dy + xy dz 6. 2eyede—2 x? + dy via” + vide = 0 7. Bate + By% dy — (09 +9 40%) de = 8. (2c! + Duy + Bez? + I) dx + dy + 22 dz =0 x +22) dx—xedy + xy de =0 10. ye (y +2) de + 2x lx +2) dy + xy (e+ y)dz =0 M1, 2G +2)dx-(rb2)dy + Qy-x42)dz=0 12.2-y)de tz (24x dy txixtylde=0. Answers : Bayes gene aig eZ 4. logayet+(xtytz)=c 5.x logz=cy Bx ty tesco 7. 1 Birtyte2ie ac Mxz=cly te) 10. xyz =cle+y 42) 12. (x yz =e (+2). 13.14, SOLUTION OF SIMULTANEOUS EQUATIONS OF THE FORM dx _dy _dz PQ R Where P,Q Rare functions of x,y, z (a) Method of Grouping Ifone variable, say z, is absent or can be cancelled in the equation £ B neaving the equation in and y only, integrate it getting /lx, y) =e,742 ATEXTBOOK OF ENGINEERING MATHEMATICS Again, if one variable, si iy x, is absent or can be neelled using (1) in the equation dy_& os = [leaving the equation in.y and z only, integrate it getting Wy, =) = i The two solutions fix, y) equation, c; and @ly, 2) = ¢, give the complete solution of the given ILLUSTRATIVE EXAMPLES de_dy a Example 1. Solve — ig ee yy mae Sol. ‘Taking the first two members, we have “= or y y Integrating, log x = logy + logc, ‘ x=cy A) Again, taking the last two members, we have x ie gaye 26" ay dz or 9-5 (Using (D) a? ey? ~ Defy? de or ozs Oop dee dz dy = or ody oe Integrating, ey = log (@- Bey) + ey or sate (<-22) +6 IQ x=cyl w or x= log (ye — 2e) — log y + 05 2) Hence (1) and (2) form the complete solution of the given equation () Method of Multipliers dx _dy_dz_Ide-+mdy+ndz Weknow that “PQ” IP+mQ+aR where Z, m,n may be functions of x, y, z or constants. If the multipliers /, m,n are auch that IP +mQ+nR=0 then Idx + mdy + nd which on integration gives olx,y,2) = ¢, 0) If it is possible to find another set of multipliors, say /', m’, n' such that VP+m'Q+n’R=0 then Idx +m’dy+n'dz=0 which on integration gives ox, y, 2) = ey AB) ‘The two solutions (1) and (2) give the complete solution of the given equation. Note. Sometimes a combination of the two methods is found useful in solving the equation.LINEAR DIFFERENTIAL EQUATIONS 743 5 dy Example 2. Solve os . x(y?— 2") y= x?) 2x — y?) Sol. Using multipliers x,y, 2 ; ads + ydy + 2dz xd + yey + edz Bach fraction = 73, Fe? Fs PF 0 s xdx + ydy +2dz=0 Integrating, x? +9? +22 =, ) % Lat Using multipliers =,—,— x'y'2 Lana Lay ae Laes days tae oni ee Buch fraction = 73a Ig Poy a Lge Layelae <9 Integrating, logx + logy +logz=loge, or xyz=c, (2) Hence the complete solution of the given equation is x* + y? +2 = c,, xyz = Cy dz 2x ay Example 3. Solve Sol. From the last two members, we have = gy & Day Baz y Integrating, logy = loge + loge, or y=c,2 Using multipliers x, y, 2, we have Beilbaad dz axdx + ydy + 2d2 ach fraction = —— =—q—g => Qaz x(x? — y? — 27) + xy? + Daz’ dz _ rdx+ydy+2d2 or oe xz x(x +9? 2?) i dz _ 2xdx + 2ydy + 2zdz he Integrating, log z + log cy or xia ye ct mex Hence the complete solution of the given equation is y=cye,2? +y? +2? = cy. TEST YOUR KNOWLEDGE Sotve de_d dx _ dy _de 1 o-2 2. = woe de _ dy a dy de & y0rs Ao Be ae z (1) (2)744 ATEXTBOOK OF ENGINEERING MATHEMATICS dz dy dz dx dy ie 3-9) dy de 2GF =e) yet ea?) ae? +9? de dy dz #Oyt 2) yet 2k) aay) 2-2) ye-* 10. xdv + ydy + 2d2 = 0, yeds + 2xdy + xyd2 = 0. Answers xaey,y seg Jog [lx +)? +27] -2e =e, xty 2 ety tem ey aye ney 6. lx emy nz ey x2 by? 424 meg ‘ ary ye seppatey ez? “6 3 genes ryt +2?= cg 8 pre aye? sey a ryt tet sey 10. xyz =e, x8 ty? 422 Cy