You are on page 1of 54
ci 2 Second-Order Linear ODEs ee LEARNING OBJECTIVES After studying the material in this chapter, you should be able to @ Know the standard form of second-order linear ODEs. © Differentiate between homogeneous and nonhomogeneous linear ODEs of second order. Understand the superposition principle or linearity principle. Know the concepts of general solution, basis and particular solution of homogeneous * linear ODEs of second order. Define an initial value problem for a second-order homogeneous linear ODE. of o @ Solve homogeneous linear ODEs with constant coefficients. © Identify and solve Euler-Cauchy equations. 4 Understand the concepts of general solution and particular solution of nonhomogeneous linear ODEs of second order. © Use the method of undetermined coefficients to find a particular solution of @ nonhomogeneous linear ODE of second order. Use the method of variation of parameters to find a particular solution of @ nonhomogeneous linear ODE of second order. 4% Discuss the existence and uniqueness of solutions of initial value p' second-order homogeneous linear ODES. roblems for 2.1 HOMOGENEOUS LINEAR ODEs OF SECOND ORDER Recall that a first-order linear differential equation is an equation which can be y'+ Pay =O) Where P and Q are continuous functions on some interva equation has an analogous form. e written in the form al J. A second-order linear differential Scanned with CamScanner 22 Maton, DEFINITION’ Second-Order Linear Dinferential Equation ‘A second-order differential equation in the (unknown) function yi) is said to be linear ig : can be written in the form y+ pay! +a@)y =") ‘ en interval J(perhaps unbounded: where p(x), 9(x) and r(x) are continuous on Sco ops ae Y ). The functions p(x) and g(x) are called the coefficients of the equation; the function r(x) on the right-hand side is called the the nonhomogeneous term. first term in the second-order linear differential equation (1) is y” 1, 4d form, However, ifthe equation begins with, say /(x)y» T Thus, we can divide by f(x) to get tae sa(l) forcing function oF Note Notice that th coefficient is 1. We call this the standart then we assume that f(x) is never zero for any ¥ € standard form (1) with y" as the first term. linear differential equation is that y” occurs only tothe ;cond-order then they do so with power | An important feature of the set th of y and y’ occur in the equation, first power, and if either or bot! only. Moreover, there are no so-call ‘A second-order ODE which is not linear is sai equation ied cross-product terms such as yy yy yy" ‘d to be nonlinear. For example, the differential yt txy2y! y= is non-linear because it cannot be written in the form (1). EXAMPLE 1 Second-Order Linear ODEs (a) The differential equation y"— Sy’ + 6y = cos.x is a second-order linear ODE because the dependent variable(or the unknown function) y and its = cos x are continuous derivatives y’ and y" appear linearly. Here p(x) = -5, g(x) = 6, and r@ functions on (0, ©). (b) The differential equation . xy" — xy! + 2y = 3x2 cos 2x is also a second-order linear ODE because it can be written in the staridard form (1) yr dyra 2y oy rad rd cos 2x where p(x) = -2/x, g(x) = 2/x?, and r(x) = 3c0s 2 e : A = are conti i contain x = 0. For example, we could take I = (0, co). SR a ee DEFINITION Homogeneous and Nonhom (DEF ogencous Lin A second-order linear ODE iS =e is called bo Yeer@y taGlyar@) is mogeneous if the forcing function r(x) on the ri ide of (1) zero (denoted by r(x) = 0), that is, r(x) = 0 for fees Tone it is: Scanned with CamScanner second-Order Linear ODEs ye nig Sunetion r(x) fe n nonhomogencou: equation (1) is nonhomogencous if the forein identically zero on 1, that is, n(x) #0 for nome x in I. For example, the differential equation vg Bye ; yey ty ott js a nonhomogencous linear ODE of second order, whereas the differential equation y" Sy" + 6y=0 is a homogencous linear ODE of second order. DEFINITION Solution of a Second-Order ODE A function y = h(x) is called a solution of a (linear or nonlinear) second-order ODE on some open interval / if h is defined and twice differentiable throughout that interval and is such that the equation becomes an identity if the unknown function y and its derivatives arc replaced by hand its corresponding derivatives, EXAMPLE 2 Show that the functions y = ¢?* and y = ¢ are solutions of the second-order homogencous lincar ODE y" Sy! + 6y=0 forall x. SOLUTION Differentiating the function y = e?* twice, we obtain yl = 20%, yi = het Substituting y = e2*, y'= 2e*, and y"" = 4e”* into the given differential equation, we obtain y" = Sy! + Oy = het — 106% + 66% = 0 This proves that the function y= ¢2* is a solution of the given differential equation. Similarly, it ean be seen that the function y = ¢3* is also a solution of the given differential equation, These two functions are not the only solutions to the differential equation, In fact, any of the following are also solutions of the given differential equation : y 2-50 y= rnc y =37e" y= 5/70 y =5e%— Ge y= 120 18e In fact, any linear combination of these solutions, such as y =3e%+20% is also a solution, Indeed, differentiation and substitution gives y" = Sy! + by = (Be + 20%)" ~ 5(3e* + Ze)! + G(307 + 2e = (1262 + 180%) ~ 5 (602 + Ge) + 6(3e% + 26%) ny o 12e2 4+ 18e8 — 30024 — 3008 + 1802 + 120% #0. Scanned with CamScanner 24 Differential Eauay, : me Homogeneous Linear ODEs : Superposition Principle a8 Consider the homogeneous second-order linear differential equation in standard form : v" + py’ + g@)y=0 | (1) where the coefficients p and q are continuous on the open interval J. i i i tion is the fact that the A particularly useful property of this homogeneous linear equat e Sum of ag two solutions of equation (1) is again a solution, as is any constant multiple of a solution, This often called the superposition principle or linearity principle. THEOREM 2.1 Superposition Principle Ify; (2) and y,(~) are two solutions of a homogeneous linear ODE (1) on an open interval Athen for any constants c, and c,, the function c, y, + ¢, y2 is also a solution of (1) on 7, Proof Since y, and y, are solutions of equation (1), we have YitP@)y, +9@)y, =9 ¥3+P@)y,+9@)),=0 By substituting y= c,y, +c,y» and its derivatives into (1), and using the basic rules of differentiation we get 2+ PCy’ + UB)Y = (ey + e232)" + PINE, + C293)" + G>EV(C hy, + )) FET 99 + CPO + erPO)Y; + e,4(@)y, + 6,409, Het + POI + 90)¥,) + 3 + PQ)YS + ga =0 This shows that y = ¢, y, + cy is a solution of equation (1). Ty simple form by means of the concept of linear combination We may state Theorem 2.1 in a ver defined as follows : A linear combination of two functions y (x) and ¥,(2) is a function of the form FO) = 440) + ep 9460) where c, and ¢, are arbitrary constants. In terms of this concept, Theorem 2.1 may be stated as follows : THEOREM 2.2 Superposition Principle (Reformulated) & For a homogeneous linear differential equation (1), any linear combination of two solutions on (an open interval Js again a solution of (1). is As an immediate consequence of Theorem 2.1 (or Theorem 2.2), we have the following corollary concerning solutions of the homogeneous linear ODE (1). (Coroliary2.3 ; | J+ Asumy, +, of two solutions y, and y, of equation (1) is also a solution (choose ¢, ==! 2. A constant multiple ky, of any solution y, of equation (1) is also a solution (choose ¢,=# and ¢, =0). ‘ = 0 is always a solution of the homogeneous linear ODE (1) Scanned with CamScanner Second-Order Linear ODEs 25 EXAMPLE 3. Itcan be readily verified that the functions y = ; the second-order homogeneous linear ODE pon aye ss ero eokioat 38 yrty=0 By the superposition principle, any linear combination ¢,cosx + ¢,sinx i : n 1C08x + c,sin.x is also a solution of th equation for any constants c, and ¢,. In particular, the li nati ‘ayia ews "y- In pi linear combination 2cosx + 3sinx is a tema a is important B emphasize that the superposition principle holds only for homogeneous linear 's; it does not hold for nonhomogeneous (linear or nonli is jonlinear) in the following examples. ') ODEs, as is illustrated in EXAMPLE 4 Consider the nonhomogeneous linear ODE yy ty=2 wa) Itcan be easily verified by substitution that the functions y = 2 + cosx and y= 2+ sinx are both solutions of equation (1), but their sum is not a solution. Neither is the function 3(2 + cosz) or 2(2 + sinx) a solution of (1). EXAMPLE 5 Consider the nonlinear differential equation yly —xy"=0 () Itcan be easily verified that the functions y= x? and y = 1 are solutions of equation (1), but their sum is not a solution. Neither is the function —x?. General Solution, Basis and Particular Solution ‘ond-order homogeneous linear differential equation The fundamental question about solution of a sec lution that contains all possible solutions. Note is concerned with its general solution —that is, a soh that the function . ye) =, YC) + eI) wae) where ¢, and c, are arbitrary constants, has the form of a general solution of equation y"+p@y' tq@y=0 (2) So the question is : can every is the expression (1) the general solution of (2) ? Thatis, bination of y, and yy? ion ; it depends on the relation between the Ify, and y, are solutions of (2), re solution of equation (2) be written as a linear com Itturns out that (1) may or may not be the general soluti solutions y, and y,. The solutions must be linearly independent. Scanned with CamScanner 2.6 Differential Equatic 7 ny The functions y, andy, are called linearly dependent on /if they are not linearly independay: Tn other words, y, and y, are linearly dependent on / if there exist constants c, and ©, not bot zero, such that 910) + eyyp(x) = 0 for all x in J. An Important Result Tivo functions y, and y, are linearly dependent on an interval 1 if ang only if one of the functions is a constant multiple of the other. Equivalently, two functions y and y, are linearly independent on an interval 1 if and only if neither of the functions is constant multiple of the other: To prove the above result, let us assume that functions y, and y, are linearly dependent. Then there exist constants c, and c,, not both 0, such that ©,9,(¢) + cyy2(x) =0 for all x in J (1) ¢ we ; . Ife, #0, then equation (1) implies y, = -* yo. Thatis, , isa constant multiple of y,. Similarly 1 ¢ . . if, # 0, then equation (1) implies yy = ~e That is, yp is a constant multiple of y,. 2 Conversely, suppose that y, = oy, on J for some constant a. Then ly, +Ca)y,=0 onl Thus, by definition, the functions y, and y, are linearly dependent. We can reformulate our definition of linearly dependent by using the concept of proportionality. Two functions y, and y, are called proportional on / if for all x on J, (@) y, =ky, or (6) » =), where & and / are numbers, zero or not. Thus, two functions are proportional if their quotient Y;/¥_ (or y2/y,) is a constant and hence the above result can be reformulated as follows : Two functions y, andy, are linearly dependentonan interval lifandonly ifthe) are, ‘proportional , Equivalently, two functions y, and y, are linearly independent on an interval I if'and only they are not proportional on I. aN For example, the functions y,(x) = e* and y,(x) = e* are linearly independent on any interval | because their quotient y,/y, = e% is not a constant, whereas y,(x) = x? and yx(x) = 5x2 af | dependent because their quotient ‘Y,/y, = 5 is a constant, | We now introduce the concepts of a general solution of homogeneous linear ODE of second ordet DEFINITION. General Solution, Basis, and Particular Solution 3 t linearly independent solutions of the second-order homogeneous linear -_Y"*P@)y' +9@)y=0 oo Scanned with CamScanner ‘Second-Order Linear ODEs a jan open interval 7. Then the general solution of (1) on Tis a solution of he Tomi YG) = 6, ¥,@) + ¥,X) a2) ee where cy, € are arbitrary constants. : ‘Any pair ¥1» ¥2} of linearly independent solutions of (1) on J is said to form a basis(or a fundamental system) of solutions of (1) on J. A particular solution of (1) on / is obtained by assigning specific values to cc, in 2). Initial Value Problem The reader may recall that an initial value problem for a first-order ODE consists of an ODE together with an initial condition y(x,) = yg. The initial condition is used to determine a value of the arbitrary constant c in the general solution of the ODE. This results in a unique solution, and hence particular solution of the ODE. We now extend these concepts to second-order ODEs as follows: An initial value problem for a second-order homogencous linear ODE consists of an ODE y"+p@y'+qey=0 together with two initial conditions YO) =k and —y"(%q) = The initial conditions are used to determine the two arbitrary constants c, and c, in the general ¢ ODE. This results in a unique solution, and hence a particular ial conditions mean that the solution curve passes through the point solution y = ¢,y, +c) of th solution. Geometrically, the in (¥q, Ay) in the xy-plane with slope &, at that point. in x are solutions of the EXAMPLE 6 Verify by substitution that the functions y, = cos x and y, ODE y" + y= 0, Then solve the initial value problem yr+y=0, y(0)=2, y'(0)=-05, SOLUTION We have yrscsinx, — y"=—cosx Replacing y by y, = cos.x and y" by y {= —cosyx in the given ODE, we obtain y" + y=—cosx + cosx =0 This shows that ¥, = cosx is a solution of the given ODE. Similarly, it can be seen that Y= sinx is a solution of the given ODE. We note that y, and y, are linearly independent on any interval because their quotient y, /y, = cotx is not a constant. Hence ‘y, and y, form a basis of solutions for the given ODE, and the general solution is y= e, cosx +c, sinx, Where c, and c, are arbitrary constants. We now use the initial conditions to find c, and c,. For this We need the derivative , Sinx +c, cosx Scanned with CamScanner Differential Equa i me 28 From this and the initial conditions y(0) = 2, »"(0) =~ 0.5, we have yQ)=c,=2 and y"(0)= 6 =-05 Hence the solution of the initial value problem is y= 2cos x — 0.5sinx. and.xe?* form a basis of the differential equation y" — 6y' +9). conditions y(0) = -1.4, y'(0) = 4.6. EXAMPLE 7 Show that e* i on that satisfies the Find also the solution that s [Dati Unie. Cay SOLUTION The given ODE is y"~6y' + 9y=0 (0) eaa] We first show by substitution that y, = e** and y, = xe3* are solutions of ODE (1). Differentiating the function y, = e** twice, we obtain yi=3e%, pt =9ee* Replacing y by e?*, y’ by 3e* and y" by 9e3* in (1), we get y" Gy’ + 9y = 9e3*— 18e5* + 9e3*=0 This shows that y, = e* is a solution of (1). We next differentiate y, = xe?* twice to obtain y5= Gxt Ne™, (Ox + 6)e3* Replacing y by xe, y' by (3x + I)e3*, and y" by (9x + 6)e>* in (1), we get y" Gy" + 9y = (9x + 6)eF— 6(3x + Ler + 9(xe*) = (9x + 6- 18x-6 + 9x)e* =0 Thus, y, = xe?* is also a solution of (1). Hence y, = e®* and y, = xe** are both solutions of (1). They are linearly independent since thet quotient y,/y, = xe**/e3* = x # constant. So e3*, xe>* form a basis of solutions of (1) and the corresponding general solution is pace +c,xe*= (c, + e,x)e* 2) We now find a particular solution that satisfies the initial conditions y(0) =—1.4, y’(0) need the derivative Y= 3(e + ere + cye* From this and the initial conditions, we have y(0) = c,=-14 yO) = 3c, + ey= 4.6 Solving these equations fore, and c,, we obtain c, =—1.4 and e, = 8.8. Substituting these Val ofc, and c, in equation (2), we obtain the particular solution of the given initial value problem y= C14 + 88x23. Scanned with CamScanner second-Order Linear ODEs 29 EXAMPLE 8 Show that x? and x-? form a basis of the differential equation x2y" +x. y'—4y=0. Find also the solution that satisfies the conditions y(1) = 11, y'(1) = 6, [Delhi Univ, GE-3, 2017] SOLUTION The given ODE is xy" txy!-4y=0 (1) We first show that y, = x? and y, = x~? are solutions of ODE (1). Differentiating the function y, 2¥? twice, we obtain Yya2m yp? Replacing y by x?, y’ by 2x and y” by 2 in (1), we get xty" + xy! — Ay =x2(2) + x(2x) —4(x?) = 2x? + 2x? 422 =0 This shows that y, = x? is a solution of (1). We next differentiate y, = x7? twice to obtain Yea-2x3, y= 6x4 Replacing y by x, y’ by -2x-3, and y by 6x~* in (1), we get x2y" + xy! Ay = x2(6x-4) + x(-2x7-3) — 40-2) = 6x2 — 2x72 4x2 = 0 This shows y, = x~? is also a solution of (1). Hence y, = x? and y, = x7? are both solutions of (1). Since their quotient y,/y, = x2/x-? = x4 # constant, they are linearly independent and hence form a basis of solutions of (1). The general solution is yHoyxttenx2, (2) where c, and c, are arbitrary constants. ‘We now find the solution that satisfies the initial conditions y(1) = 11, y'(1) =—6. We need the derivative y= 2c,x-2¢,x-3 From this and the initial conditions, we have yl) =e, #ey= 11 y"(1) = 2¢, - 2¢)=-6 Solving these equations for c, and c,, we obtain c, = 4, c) = 7. Substituting the values of c, and cy, in (2), we obtain the solution of (1) satisfying the initial conditions as YQ) = 4x? + 7x, Namnncnsisy a EXERCISE2.0 00 1. Show that e* and e-* form a basis of solutions of the ODE y” — y= 0. Then solve the initial value problem y"-y=0, y(0)=6, y'0)=-2. 2. Show that cos x and sinx form a basis of solutions of the ODE y" + y = 0. Then solve the initial value problem y+ y=0, yO)=3, y')=-0.5. Scanned with CamScanner 2.10 Differential Equay, ns 3. Show that e* and e~2* form a basis of solutions of the ODE y" +y’—2y'=0. Then soiye initial value problem . y" ty! —2y=0, (0) =4, (0) =-5. 4. Show that e4* and e~** form a basis of solutions of the ODE y” ~ 16y= 0. Then solve y, initial value problem y"—16y=0, y(0)=3, yO) =8. 5. Show that x3 and x3 form a basis of solutions of the ODE x?y"— 7xy' + 15y=0, The solve the initial value problem x2y"— Ixy" + 15y=0, y(I)= 04, y= 1. 6. Show that cos 5x and sin 5x form a basis of solutions of the ODE y" + 25y=0. Then solve the initial value problem y"+25y=0, y(0) = 0.8, y"(0) = 6.5. 7. Show that e~9* and xe~°5* form a basis of solutions of the ODE y" + y’ + 0.25y=0, They solve the initial value problem y" ty! +0.25y=0, y(0)=3, y"(0) =-3.5. of solutions of the ODE y" + 2y’ +2y=9, 8. Show that e~*cos.x and e~*sinx form a ba Then solve the initial value problem pre dy! +2 =0, y(0) y'0)=-1. f solutions of the ODE y” + 2y' + y= 0. Then sole 9. Show that e~* and xe~* form a ba: the initial value problem y" + 2y'+y=0, y(0)=4, y'(0) ANSWERS 2. y =3cosx ~0.5sinx =25et* +0,5e4 = 0.8cos 5x - 1.3 sin5x Ly =2e*+4e-* Bey =et+3e—2* 5. y =0.5x3-O.1x5 6. Tey =(3-2xje705* 8. y =e-*cosx 9. y =(4-2x)e* 2.2 HOMOGENEOUS LINEAR ODEs WITH CONSTANT COEFFICIENTS In this section we learn how to solve second-order homogencous linear ODEs with constant coefficients, that is, equations of the form eal) y"+ay'+by =0 f where a and 6 are constants. Scanned with CamScanner second-Order Linear ODEs ‘To find the general solution, we first look for a single soluti solution of the form y= e**, Substituting y = e** and its dehaie yi =he andy" =22ehe jnto the differential equation (1), we obtain 211 4.(1). We begin by assuming a ives Me* + a(e®) + b(e*) =0 ies eM (2 + ah +b)=0 Because e”* is never zero, we see that y of the equation += will be a solution of Eq.(1) precisely when is a root B+ak+b =0 @ Equation (2) is called the charaeteristic equation of the differential Eq,(1). Our problem, then, is reduced to the solution of this quadratic equation. Observe that Eq,(2) is formally obtained from Eq(1) by merely replacing y" by 2, y' by, y by 1. Eq.(2) has two roots, given by the quadratic formula : Ay dy = ‘These roots may be real and distinct, real and repeated, or complex conjugate according as the discriminant a? — 4b is positive, zero, or negative, respectively. Case 1. Distinct Real Roots If the discriminant a? - 4b > 0, then the characteristic Eq.(2) has 2 distinct real roots A, and 2). Moreover, cur derivation shows that the functions de dar yy =e andy, = & are solutions of equation (1). These solutions are defined for all x and their quotient ype eye = e1** x constant (since 2, #22). Hence e*, e™* forma basis of solutions of (1) on any interval and the corresponding general solution is Re y =ce™ +e,e Where c, and ¢, are arbitrary constants. EXAMPLE 9 Find a general solution of the differential equation y" + 2y" — 8y= 0. SOLUTION The characteristic equation is te 22420-8 =0 or (A44)(Q-2)=0 Ithas two distint rel roots 2, = ~4, y= 2:-Thus, the general solution of the given ODE is y =ce* + 0,07. EXAMPLE 10. Find a general solution of the ODE 10y"~7y" + 1.2y= 0. SOLUTION Dividing by 10 the given ODE can be put in standard form y"—0.7y' + 0.12y=0 Scanned with CamScanner 212 ‘The characteristic equation is 22-0.7.40.12 =0 or It has two distinct roots 2, = 0.4, 2, = 0.3. Thus, the genera y = ce + 0,69. Differential Equation, f q (2 - 0.4) — 0.3) = 0 | solution of the given ODE jg Case 2. Real Double (or Repeated) Root 2 4h ig zero, then the characteristic equation has a real double root If the discriminant a R= dy = hy = al? and hence we get only one solution, namely, y = e*, However, it a jedeyrenneires ysreds also a solution. Since the solutions e** and xe** are not proportional, hey a basis. Thus, iy the case of a double root 2 of (2), a basis of solutions of (1) on any interval is yy = e+ ang Yq = xe*, The corresponding general solution is y =cyerF + e,xe™*=(c, +e,x)e™* where ¢, and c, are arbitrary constants. EXAMPLE 11 Find the gencral solution of the differential equation y"” — 6y’ + 9y = 0. SOLUTION The characteristic equation is . -6249=0 & (2-37 =0 Ithas the real double root A = 3. Hence the general solution is y =cye* + egxe™ = (c, +e x)e* where ¢, and c, are arbitrary constants. EXAMPLE 12 Find a general solution of the ODE 4y"~20y' + 25y = 0. SOLUTION Dividing by 4 the given ODE can be put in standard form y"— Sy" + 6.25y=0 ' The characteristic equation is 2-524+625 =0 or (2-25)?=0 Ithas the real double root 4 = 2.5. Thus, the general solution of the given ODE is Y =e, + exe, Case 3 Complex Conjugate Roots Ifthe discriminant a? — 4b <0, then the characteristic equation (2) has complex conjugate 1005 yp dy = AE Mat ade a, Vab~ a? : » eo A 2 Mend, Scanned with CamScanner yp-Order Linear ODES apie te roots a8 follows: 213 a hy =F ae ig = O-i0 jn the cave of real roots, the exponential functions y, = Pcl equation (1). Thus, the general , ¥F Jae * and y, solution is are solutions of hye ec, and c, are arbitrary constants. However, the solutions bese pena " 7, the solutions defined by e+ 9) and ela-ie)z we ope S00 of the real variable x. It is de: es ‘sirabl 7 . iabependent solutions. This can be accomplished by « to replace these by two real linearly using Eular’s formula : = : 609+ isin agpich holds for all real 0. Using this, we have eye 108 + ene ta)e oye 19% + cett e-tox = [eel + ce-ior] = e%[e\(coswx + isinox) + ¢,(cos@x—isinox)] =e" [(c, + c,)cosex + i(c, —c,)sinox] =e [Acoswx + Bsinwx] where A= cy + ¢,, B= i(c, ~ c,) are two new arbitrary constants. Thus, the general solution corresponding to the conjugate complex roots o + i is y(x) = e"*[Acos ox + Bsin ox). EXAMPLE 13. Find the general solution of the differential equation y” ~ 4y’ + 13y=0. SOLUTION ‘The characteristic equation is 92-42.413=0 or (2-2)2+9=0 = 2-2=43i or 2=2431 ‘Thus, the characteristic equation has the complex conjugate roots 2 * 3i. Hence the corresponding gencral solution is y =e[Acos 3x + Bsin 3x] Where A and B are arbitrary constants. . EXAMPLE 14 Find the general solution of the ODE y"~y' +2.5y=0. SOLUTION The characteristic equation is 12-4425=0 of (2-0.5)24225=0 = h-05=#157 onjugate roots 0.5 + 1.5i. Hence the or 4=05415i Thus, the characteristic equation has the complex c ‘corresponding general solution is y = e9Sx[Acos 1.5x + Bsin 1.52), Where A and B are arbitrary constants. Scanned with CamScanner 214 EXAMPLE 18. Find a general solution SOLUTION The characteristic equation is M+ 2A +40 or (ar aye 2.5600 ation has the complex conjugal ‘ofthe differential equation y"" 2Ayl bh AOy 0, AHLZHLG OF Rm =12 416, ¢ roots = 1.24 1.61, Hen Thus, the characteristic equ 1 the corresponding general solution i ye entt[dcos Lox + Bsin 1.6.x), where 4 and B are arbitrary constants. | EXAMPLE 16 Find the general solution of the differential equation y" + @?y = 0. SOLUTION The characteristic equation is 22-42 = 0, Ithas the complex conjugare roots + «yj, Hence the corresponding general solution is | y = Aoos ox + Bsin ox where 4 and B are arbitrary constants. Initial Value Problem : Some Illustrations yreyt-2ys0, yQ=4, "O= SOLUTION Step 1 General Solution The characteristic equation is 2242-2=0 fe, 224+2A-2-2=0 on, (A-1)A+2)=0 Ithas two distinct real roots 4, = | and Ay = -2. Thus, the general solution is y =ce" + c,0-%* eal) Step 2 Particular Solution : ‘We now use initial conditions to find constants c, and cy. We need the derivative Y= ee*—2e,e* From this and the initial conditions, we obtain | W(0) =e, +e, =4 3(2) ¥'(0) =e, 2c, @ Solving Eqs.(2) and (3) for c, problem has the solution 1 and ¢>, we obtain'c, = | and c, = 3, Thus, the given initial value y Fer + 3e7%, EXAMPLE 18 Solve the initial value problem | 2y"— By" 2y =0, (0) = 3/2, y'0) =-1/8, Scanned with CamScanner aunt Onder Linear ODES solution phe character 2S pt wt Solution ie equation is WeMaWO a EAHA rrhas wo distinet real rots Ay 1/2 and Ay = 2, Thus, the genoral solution ly ym oet? teem 7 wl) sep 2 Particular Solution We now use initial conditions to find constants c, and cy, Wo need the derivative ale 9 yl = sage eye?" From this and the initial conditions, we obtain yO) =e, Foy @ 32 (2) ; 1 YO) = Ze, Heys -t Q) 3 as ' From Eq.(2), cy = 5 > 4: Substituting this value of c, in Eg.(3), we obtain or Thus, the given initial value problem has the solution 5 p12, Las yaqos ge tial value problem 90) = 3, YO) = 35. EXAMPLE 19 Solve th yt yt + 0.28y = SOLUTION Step 1 General Solution The characteristic equation is q+ 402520 o (AOSP =O Ithas the real double root 4 = -0.5, Hence the general solution is y =(c, tere Step 2. Particular Solution We now use initial conditions to find constants c, and ¢,, We need the derivative yh eye O58 — 0,5(c, + pte et w(t) Scanned with CamScanner Differential Equation, : 2.16 From this and the initial conditions, we obtain y@) e,"3 (2) yO) = 0.5, Fe = 35 +.) hie imply that cy 3 an ¢y = =2, Thus the partite solation of the given initial value problem s = (B= 2x)er0", EXAMPLE 20 Solve the initial value problem prt ayrry 0, yO) = 5, y= SOLUTION Step 1 General Solution ‘The characteristic equation is Men+r1sd0 @ GtlP=0 1. Hence the general solution is thas the real double root 4 = y ye" (1) Step 2 Particular Solution We now use initial conditions to find constants ¢, and c,. We need the derivative y. en* From this and the initial conditions, we obtain »(0) (2) »Q) yO) 2 : which imply that cy = $ and c, = 2. Thus, the particular solution of the given initial value problems y =(+2xe". EXAMPLE 21 Solve the initial value problem y"+Ody'+9.04y =0, y(0)=0, y'(0)=3. [Dethi Univ, GE-3, 2018] SOLUTION Step 1 General Solution The characteristic equation is +04K49.04=0 or (A+0.2)2+9=0 => A+0.2=43i or 2=-0.243i Thus, the characteristic equation has conjugate complex roots -0.2 + 3i. Hence, the general solution is y =e~"2*[4 cos3x + Bsin3x], where A and B are arbitrary constants. Step 2 Particular Solution ‘We now use initial conditions to find constants 4 and B. The first initial condition gives »(0) =A=0 Thus, we are left with the expression y = Ben sin3x To apply the second initial condition, we need the derivative > y' = BE-0.2e~°2*sin3x + 3e-02*¢953x) Scanned with CamScanner second-Order Linear ODEs ~qhe second initial condition gives . 247 am YO) =3B=3 => Bay Hence the particular solution of the given initial value Problem is Y =e 2sin3x, EXAMPLE 22 Solve the initial value problem »"~ 6y' + 25y=0, (0) =-3, SOLUTION Step 1 General Solution The characteristic equation is ~OR+25=0 or Q-3P+16=0 = A-3=44i of 2-344) Thus, the characteristic equation has conjugate complex roots 3 + 4i. Hence, the is y') general solution y = e**[Acos4x + Bsin4x], ea) where A and B are arbitrary constants. Step 2 Particular Solution We now use initial conditions to find constants A and B. The first initial condition gives y0) =A=-3 To apply the second initial condition, we need the derivative > y! = e*(-4Asin4x + 4Bcos 4x) + 3e?*(Acos4x + Bsin4x) = e*[(34 + 4B)cos 4x + (3B— 4asingy] The second initial condition gives > y'(0) =344+4B=-1 Substituting A = —3 in the last equation, we find B = 2. Substituting the values of A and B into Eq,(1), we obtain the particular solution of the given initial value problem : y = e3*(-3cos4x + 2sin4x). Recovering a Differential Equation from Solutions We can also work backwards using the results above. That is, we can derive a second-order homogeneous linear ODE with constant coefficients that has a given basis of solutions. Let us consider some examples. EXAMPLE 23 Find second-order homogeneous linear ODE with constant coefficients that has the basis e?*, e3 of solutions. SOLUTION To the given basis e**, e~>* of solutions, there corresponds the characteristic equation. Q- 2) +3) =A? +A-6=0 Hence, the corresponding ODE is y"+y'—6y=0. iii : Scanned with CamScanner a Differential Equation, EXAMPLE 24 Find second-order homogeneous linear ODE with constant coefficients theta the basis 1, e~* of solutions. SOLUTION To the given basis 1, e~** (ie, e°*, e~**) of solutions, there correspond the characteristic equation AQ. +2)=224+220=0 Hence, the corresponding ODE is yu t2y'=0. EXAMPLE 25 Finda second-order homogeneous linear ODE with constant coefficients that has the basis e8*, xe°** of solutions. SOLUTION To the given basis e¥5*, xe" of solutions, there corresponds the characteristic equation a-By Hence, the corresponding ODE is y"— 23 y'+3y=0. -2N3R+3=0 EXAMPLE 26 Finda second-order ‘homogeneous linear ODE with constant coefficients that has the basis e@!* 2, ¢~(1 +24) of solutions. SOLUTION To the given basis e! +2), 9-(1 + 2i)r of solutions, there corresponds the characteristic equation Q+1-2)A+1 +2i)=(A+1)?- (22 =2242045=0 Hence, the corresponding ODE is y" +2y' + 5y=0, EXAMPLE 27 Find a second-order homogeneous linear ODE with constant coefficients that has the function y(x) = e* cos 2x as a solution, SOLUTION Since ¢*cos 2xisa solution, 1 +2’ must bea root of the characteristic equation. We know from algebra that roots of a polynomial equation with real coefficients occur in conjugate pairs. Therefore, 1 - 2 must be the other complex root of the characteristic equation. These roots correspond to the characteristic equation @-1-2)Q -142)=A-12-n2=2 45 =0 Hence, the corresponding ODE is y"—2y' +5y=0. iii sini Scanned with CamScanner EE Se SEE TEE TRE TEE TES ESE S TREE TESERE SE SESEE IE SECEETSSERESRTEREOOEIESIS® secont-Orer Linear ODES 23 DIFFERENTIAL OPERATORS 219 dy Incalculus, We often use the capital letter D to denote differentiation ; that i The ares nbo| Dis called the differential operator because it transforms a given (differentiable) function ip its derivative. For example, Dsinx = (sin $x, Dx? + Sx + 1) = Bx? + Sx + LY! gc + 5. Higher-order derivatives can also be expressed in terms of D in a natural way : d d"y polynomial expressions involving D such as D- 2J, D? + 3D + 21,x2D? + 4xD + 51, etc., where Tis the identity operator defined by Jy = y, are also differential operators. In particular, the expression L=P(D)=D*+aD+bI is called the second-order differential operator with constant coefficients. ‘Animportant property of second-order differential operator £ is that itis linear in the sense that if Ly and fw exist, then for any constants a and b, we have L(ay + bw) = aLy + bLw Using the second-order differential operator L, a second-order homogeneous linear ODE with constant coefficients y" tay" +by =0 can be expressed as Ly = P(D)y=(D? + aD + bly =0 For example, the second-order homogeneous linear ODE y"=2y"+5y=0 can be expressed in terms of differential operator D as (D?-2D+ sly =0. Note Notice that P(D) can be treated just like an algebraic quantity. In particular, P(D) can be factored in the same way as the characteristic polynomial P(A) = A? + ad + b. é "EXERCISE 2.2” Finda general solution of each of the following ODEs. Ly" +5y/-6: 2. yt + 8y' + 16y=0 3. y"-2y'—5.25y =0 4, y"+2y'+5y=0 S.y"+ Any! +4n2y=0 6. 100y" + 20y"— 999 =0 7. y" + 26y' + 1.69y=0 8 y"—3y=0 9% y" + y"—0.96y=0 10, y”-225y=0. I, (D?+2D+41y=0 12, (D2 + 6D+ 13/)y=0 Scanned with CamScanner MPM EQuatio, | ns Solve each of the following initial value problems. =0, (0) =2, yO) =0 14, yp" -99 = 0, 90) =-2, y'(O)= ay ty=0, P=1, "O=-! 16, yr t4y' + 5y=0, VO) =2, y'O)=_s 17. Loy" + Sp" +.0.625y = 0, y(0) = 2s y'@)=-45 18, (D2? -2D-24Ny=0, y(0) = 0 y'(0) = 20 19. (10D? + 18D + 5.6/)y=0, ¥(0) 4, y'()=-38 ) = -k. 20. y" + 2ky! + (2 +02) = 0, YO) = 1b vO Find an ODE y" + ay" + by = 0 for the given basis. 2 en, e* 22. e, e** 23, e738, xe ¥ 24, elt), etd 25.1, 26. 35%, e}5* ANSWERS 2 y=(e, + exe" 4. y= e™[Acos 2x + Bsin 2x] v=ceS +e,er ce + ce ec euttoenkt 3 yacer +ce 6. y= ce0% + e-H* vx 5. y= (c, + exe + oe T.y=(c, + qxen!* 9. y= c,e98 + eye! 11. y= e™[Acos J3x + Bsin J3x] —Sy2r, Ao3x 13.y= Sets be 31 _ 3938 15. y=(1-2x)e* / 16, y= e-?*(2cosx — sin x) 17. y= (2—4x)en025* 18. y= 20% — 2e-4* 19. y= 180-048 +2.2¢-14r 20. er cos@x 2. y"—6y'-Ty=0 22. y"=2y'—24y=0 23, y" + 2.6y' + 1.699 =0 24, y"+2y'+2y=0 25, y"-+5y"=0 26. y"~ 2y'-5.25y=0 2.4 EULER-CAUCHY EQUATIONS The Euler-Cauchy equation is an ordinary differential equation of the form x? y"+axy! + by=0 (1) where the coefficients a and b are constants, The i . I 7 coeffi ii J er th cenit ad const The concn of he tm yi |W Scanned with CamScanner second-Order Linear ODEs qhe trick for solving this equation is to Ty 2 solution of the form cs ¥ =x" sere m i fo be determined. The idea is similar to that for homoge gen constant coefficients. Differentiating this function, we ous linear ODEs with obtain YS mxBh y= mim 1)x8-2 x" and its derivatives y’ = mx"-! and > zs m(m = 1)x"-2j . m(m —1)x" + amx™ + x= =0 DP" into Eq(1), we get or, (m(m — 1) + am +5)x™=0 x" is a solution of the differential equation (1) if and only if m is a root ofthe auxili m(m-1)+am+b=0 which simplifies to m? +(a-1)m+5=0 2) There ave three different cases to be considered, depending on whether the roots of this quadratic equation are real and distinct, real and equal, or complex. In the last case the roots occur as a conjugate pair. Case 1. Distinct Real Roots If the roots m, and m, of the auxiliary equation (2) are real and distinct, then the solutions yy =x" and are linearly independent because their quotient is not constant. Hence x”, x"? form a basis of solutions of (1) for all x for which they are real. The corresponding general solution is yaar t+ ex"? where c, and c, are arbitrary constants. EXAMPLE 28 . Solve the Euler-Cauchy equation x?y" — 2xy’-4y=0. SOLUTION Comparing the given equation with the standard Euler-Cauchy equation by=0, We find a=—2, 6 =—4, Hence the given Euler-Cauchy equation has the auxiliary equation m2 —3m—4 =(m+ 1)(m-4)=0 1 and m, = 4. So the general solution is ye erttext Its roots are m, i ji 2 + xD—ADy=0, where EXAMPLE 29 Find a general solution of the differential equation (2D?+xD-4DyY d D= [Delhi Univ. GE-3, 2016] Scanned with CamScanner | 2.22 SOLUTION The given ODE is an Eular-Cauchy equation xy" +xy'—4y=0, Differential Equa _ ‘ony with a= 1, 6 =—4. The auxiliary equation is m?+(a-1)m+b=0 ie, m-4=0 => m=2,-2 Thus, the auxiliary equation has two distinct real roots 2 and —2. Hence the general Solution Of the given ODE is Y= Ox? + ox, EXAMPLE 30 Find general solution ofthe differential equation (x?D? + 6x +6/)y=9, at D= 4 Welhi Unix, GE-3, 2019 SOLUTION The given ODE is an Eular-Cauchy equation x?y" + 6xy' + 6y=0, with a = 6, b= 6. The auxiliary equation is m?+(a-1)m+b=0 ie, m+5m+6=0 = or (m+ 2)(m+3)=0 => ms=-2,-3 Thus, the auxiliary equation has two distinct real roots ~ 2 and ~3, Hence the general solution of the given ODE is . mex + ox-3 Y= Cx? + cyx-3, EXAMPLE 31 Solve the following differential equation (x D? + 4D) y = 0. Also find the solution satisfying y(1) = 12, y'(1) = -6. (Dethi Univ. GE-3, 2018] SOLUTION The given ODE can be written as xy"+4y'=0 > xy"+4xy'=0 (provided x #0)” This is an Eular-Cauchy equation with a= 4 and b= 0, The auxiliary equation is m? + (a-1)m+b=0 ie, m+3m=0 => m=0,-3 Thus, the auxiliary equation has two distinct real roots 0 and —3. Hence the general solution of the given ODE is i YRC, tex} We now find the particular solution satisfying the conditions y(1) = 12, y'(1) = 6. We need derivative y= -3ex-4 From this and the initial conditions, we have | YQ) = ec, +0, = 12 Y'Q)= -3c,=-6 Scanned with CamScanner Order Linear ODEs second: 2.23 = 2. Substituting these values in the tial value problem solving these equations for c, and c,, we obtain ¢, = 10 and, general solution, we get the particular solution of the given ini y= 1042273, case 2. Real Double (or Repeated) Root ifthe auxiliary equation (2) has a real double root m(that is, m= m, = linearly independent solutions are y, =x" and y, i)» then the two corresponding =x" In.x. The corresponding general solution is Y=Ex™+e,x"Inx, EXAMPLE 32 Solve the differential equation : 4x2y" + 8xy'+y=0, SOLUTION The given differential equation can be written in standard form of the Buler-Cauchy equation ° y's dys Ty =o Here a= 2, b= 1/4. Its auxiliary equation is m? +(a-1)m+b=0 1 ie, m+Q-Mm+5=0 or Ant +dm+1=0 or (2m+1)=0 ‘Thus, it has the real double root m = —1/2. Hence the general solution for all positive x is y =e,x-¥2 U2 EXAMPLE 33 Solve the initial value problem: x?y" + 3xy" Fy =0, y= 4 y"(1) SOLUTION Step 1 General Solution Inx. xl + ex (Dethi Univ, GE-3, 2016] The given ODE is an Euler-Cauchy equation with a= 3, b= 1. The auxiliary equation is m?+(a-1)m+b=0 ie, m?+2mt+1=0 or (m+l)=0 Ithas the real double root m =—1. Hence the general solution for all positive x is y=(c,te,Inx)x Step 2 Particular Solution We now use initial conditions to find constants c, and c,. We need the derivative -(c, + @Inx)x? + cx? y From this and the initial conditions, we obtain yd)=e,=4 yi) =e + ey =-2 Scanned with CamScanner Sas, sven initial vane... fat Solving these equations, we find e, = 4, cy = 2. Thus, the given initial value probjen, bn m solution th ys(4t2inaxx! Case 3. Conjugate Complex Roots jomial equation with real coefficients occur in go. We know that the complex roots of a polym ies ‘ ; pairs. Thus, if A = a + i@ is a simple root of the auxiliary equation (2), 80 is the cong Z =a—io, and the two corresponding linearly independent solutions are Bae yy =x%eos(@lnx) andy, =x"sin(oInx) The corresponding general solution is y =x®[Ac0s(@ Inx) + Bsin(@ Inx)). EXAMPLE 34 Solve the Euler-Cauchy equation : x?y" + 0.6xy’ + 16.04y = 0. SOLUTION The auxiliary equation of the given ODE is m? + (0.6 — 1)m + 16.04 = m? — 0.4m + 16.04 = 0 Using the quadratic formula, the roots of the auxiliary equation are 0.4 + 0.16 — 64.1 . i oe 64.16 0458 0224 Thus, the general solution of the ODE is y =x°[4cos(4Inx) + Bsin(4Inx)]. EXAMPLE35 Solve the initial value problem : 2x2y"txy'-y=0, y(=1, y'(I)=2. SOLUTION Step 1 General Solution Dividing by 2, the given ODE can be written in standard form of the Euler-Cauchy equation rw kd xyr pty lye Wr oy 0 Here a= 1/2, 6=-1/2. Its auxiliary equation is mm +(a—1)m+b=0 2 1 1 m+ (S-1m-t =0 or 2m?—m-1=0 or (2m+1)(m-1)=9 ie, Tthas evo distinct real roots m, = 1/2, m= 1. Thus, the general solution of the ODE i 1 Yr exl? +e x=q— + ox vx 2° | Where c; and c, are arbitrary constants, Scanned with CamScanner order Linear ODEs aos ep 2 Particular Solution ow use initial conditions to find constants ¢, and c,. We need the derivative 1 we . 32 gx? + ec y srom his and the intial conditions, we obtain yO) =e, 4eg= 1 1 yi) =-34 + solving these equations, we find c, = 2/3 and ce, = $/3. Thus, the particular solution of the initial value problem is EXERCISE 2.3 Find a general solution of each of the following ODEs. 1, x2y" + 2xy'- 6y=0 2. x2y"—Txy' + 16y=0 3. x2y" + 0.5xy" + 0.0625y = 0 4. xty" xy" +4y=0 5, 4x2y" + 4xy'-y=0 6. x2y"+3xy' +y=0 7, xty" —2xy! + 2.25y=0 © 8. x2y" + Txy'+ 9y=0 9, x4y" + 3xy! + 2y=0 10. x*y"-0.75y=0 Solve each of the following initial value problems. UL xy" 6y=0, p= 1, yr") =0 12, x2y"— 4xy' + 6y=0, p(1)=1, y= 0 13, x2y"+ 3xy' +y=0, yl) =4, y"(1) =-2 14, x2y"— Ixy’ + 16y =0, yl) =4, "1 =-2 15, (x2D2 —2xD + 2.25Dy =0, y(1)=2.2, y'()=2.5 16. (x2D2 + 1.5xD - 0.5) y= 0, yl) =2, y'(Q) =—L. Leyex? +e,x 3. y= (e+ ey Inx)x95 5. y= vx + /Vx 6. y=(c,+e,Inx)x* 8. y=(c, +einx)x> 2. y=(e, + cInx)x* 4. y = Acos(2 Inx) + Bsin(2 Inx) Tey =(c, + ey Inx)x!5 [Acos(In x) + Bsin(In x)} 10. y=e,xt5 + 0px y= Scanned with CamScanner Differential Equation, 23x2-2x7 I y= 284 3x? 12,953 3.22 14. y=(4-18Inx)x* 13. y= “| Z 3. y= (4+ 2Inx)x care 15, y= (2.2-081nx)x!5 [D UNIQUENESS OF SOLU blish the general theory of second-o TIONS. WRONSKIAN 2.5 EXISTENCE ANI In this section we shall estal yt + pay! + q@)y = are continuous on some Opel ; general solution of (1) as well as the uniqueness ial conditions rder homogeneous linear ODE; . (1) n interval J, Our main concern will be the pibere pts) ond 90) of the initial value problem existence and form ofa pil OF consisting of such an ODE with two initi : WE) = hy ED =A with given values xo, and k,. The following Theorem (proof omitted) states that such an initial value solution that is unique. THEOREM 2.4 Existence and Uniqueness Theorem for Initial Value. problems Ifp(x) and q(t) are contimious on some open interval / containing the point xo, then the initial value problem problem always has a y"tp@y' + q@)y=0, yO) =k Wo) = ky has a unique solution p(x) on J. Linear Dependence and Independence of Solutions. Wronskian Recall from Section 2.1 that two functions y, and i i i Rca ti ions y, and y, defined on an interval / are linearly independent e710) +e,9,2)=00nt > ¢,=0, e=0 The functions y,, y, are li i ist constanis « ee 1» ¥2 are linearly dependent on / if then exist constants ¢, and c,, not both zero, Mec N10) + eyy2(x) = 0 on I was also observed in Section 2.1 that two functi s i Si : ; only if they are proportional on J that sn? 12S linearly dependent on /ifard either =) = where k and / *=ky, OF y,=1y, forall ver, ia are numbers, zero or not. In contrast, j : ions are not proportional. in the case of linear independence thes? Linear independence of solutions i lutions is crucial fc os to ha iteric . . ial for obtainin; . p ve a criterion for it. For this we introduce the foll 8 general solutions. Thus, it would be good criterion. lowing definition that is relevent for stating. Scanned with CamScanner second-Order Linear ODEs a GERINITION Wronskian of Solutions : and y, be two solutions of second-order homogeneous linear ODE y" + p@)y' + q@)y=0 ‘the Wronskian of y,,y2, denoted by IV(y,, y2) i defined as the determinant n WI) =| + vA Let Note It is important to emphasize that the Wronskian is a function of x that is determined by two solutions y, J. of ODE (1). For this reason we sometimes use the notation I7(y, ¥3)(6) t0 emphasize that Wronskian is a function of x. When there is no danger of confusion, we'll shorten the notation to W(x). ‘The Wronskian can be used to determine whether or not two solutions of (1) are independent as is stated in the following Theorem. THEOREM 2.5 Linear Dependence and Wronskian Lety, and y, be two solutions of the second-order homogeneous linear ODE y" + py" + g@)y=0 (1) where the coefficients p(x) and q(x) are continuous on an open interval /. Then the following statements are equivaluent : (@) yy,y2 are linearly dependent on J. (1) Wry y) = 0 on L. 0) WO I)Co Proof (a) => (b) We assume that y, and y, are linearly dependent on /. Then they must be proportional, that = 0 for some nl. her (a) y= ky, or (b) yy =/y, forall eT and /. for some constants Ify, = ky, then Wop yyy 5 — ya ky =O ont Similarly, if y, = /y,, then Wy rd =ryy— yw aM yi 0 on Thus, in either case, Wronskian is identically zero on J. (6) = (©) Obvious. (©) = (a) We assume that I7(, y3) = 0 for some x9 in 1. We must show that y, and y, are linearly dependent on J. We consider the homogeneous linear system of equations in the unknowns Spe a=(2) 1 y1(%0) + €2 ¥2(%o) = 0 c1¥, o) + 292(%0) = 0 Scanned with CamScanner Differential Eaquay, t ‘Ong 2.28 This system in matrix form is yo) 2640] fer -[] co} LO. (3) c 9,9) 9360) oO y system (2) (or system (3)) has a nontrivial solution if and only ig fficient matrix Y is zero. But this determinant is simply the Wrong e W = 0. Hence the system (2) (or system (3)) hag a From linear algebra the .d, by assumption, system (2 not both zero, satisfying linear equations in SYstem, determinant of the coe! evaluated at x = Xqy an nontrivial solution. Thus, there exist c, and cp, : (Q). With these numbers c,, ¢, we introduce the function pC) = 19408) + 2320) By the superposition principle this function is a solution of (1) on J. From (2), we see that it satisfies the initial conditions y%) = 9, ¥'&) = 0, of (1) also satisfies the same initial conditions. Since the However, the trivial solution y* solution of initial value problems is unique (Theorem 2.4), we have ysy* ie, oy tey,=0 onl Now since c, and c, are not both zero, the solutions y, and y, are linearly dependent on J. This completes the proof. The following corollary gives a useful characterization of linear independence using the conceptof Wronskian. Corollary 2.6 Linear Independence and Wronskian : Two solutions y, and y, of (1) on J are linearly independent if and only if their Wrondlian |W (.93) is never zero on I: ee En eee EXAMPLE 36 Consider the ODE B yey Oy y substitution, it can bi 7 F , a" Recaien is e shown that y, = cos@x and y, = sin@x are solutions of (1). Their cosox — sinax |, -osinox wcosax no” W0vy) =| B Thus, it follows fr o=0,we reenter i a and y, are linearly independent if and only if © 0.Fo independent solutions never contains Hapa lneet dependence because a pair of lineaty = @cos*wx + wsin’ox =O EXAMPLE 37 Consider the ODE y= dy’ +y=0 By substitution, it ” (-%, »). By a eee ca and J = xe* are solutions of (1) on the ine ae) 2y Pepetin nearly independent, This fact can also be see? ¥Y Scanned with CamScanner thy his second-Onder Linear ODEs e* wader] From Corollary 2.4, these solutions are linearly independent, 2.29 et Nem yyte =o nv WO =|» ” Which is never zero ¥, E38 Fi gencous linea EXAMPLE 38 Finda homogeneous near ODE ‘ond-order for which the functions ¢2* ef ae solutions. Also show linear independence by considering their Wroruky ons ee and n, SOLUTION To the given pair e*, e* of solutions, there Corresponds the characteristic equation Q-DA-I)= 2-3-4220 ee Hence, the corresponding ODE is vy" -3y' + 2y=0 The Wronskian is ex 202 ot wove wove =e! 2 elt ny) WO¥) = | Thus, the given solutions are linearly independent. EXAMPLE 39 Find a homogeneous linear ODE of second-order for which the functions e** and xe** are solutions, Also show linear independence by considering their Wronskian, SOLUTION To the given pair e*, xe of solutions, there corresponds the characteristic equation QW? =22-2k k=O Hence, the corresponding ODE is yh = 2ky' + Ry =0 The Wronskian is nS: Jhx xe x nov Thus, the given solutions are linearly independent, EXAMPLE 40 Find a homogencous linear ODE of second-order for which the functions cos rx and sin zx are solutions. Also show linear independence by considering their Wronskian, é kek* (hee + Ne* = (ki + Ne — brett = 2k 4 9 wovs9-| SOLUTION To the given pair cos 7x, sin7x of solutions, there corresponds the characteristic equation, Q-n)A+ni) =A +27=0 Hence, the corresponding ODE is yrtry=0 The Wronskian is costs SINEY | = neostnx + nsin?nx =n #0 -msinax mCosmx| nom Thus, the given solutions are linearly independent. MY WO yr.) | Scanned with CamScanner a — WMatio, ond order for which the functions x3 idering their Wronskian, and x. [Delhi Univ. GE-3, 25, +2014 2.30 EXAMPLE 41 2 are solutions. Als ous linear ODE of seco cd a homogenc 0 idence by consi show linear indepe 2 of solutions, there corresponds the auxiliary equation 6=m(m-1)-6=0 SOLUTION To the given pair (m= 3) (+2) =P =m Henee, the ODE (an Eular-Cauehy equation) is xy" = 6y=0 The Wronskian is 3 M2 wn v2 Wy) = 3x? -2x Thus, the given solutions are linearly independent. AMPLE 42. Find a homogeneous linear ODE of second order for which the functions x? ang dependence by considering their Wronskian, E In.x( > 0) are solutions, Also show linear in [Delhi Univ. GE-3, 2017 2019 x SOLUTION To the given pair x~3, x~? Inx of solutions, there corresponds the auxiliarly equation (m +3)? =m? + 6m +9 = mP—m-+ Tm +9 = mm — 1)+7m+9=0 Hence, the ODE (an Eular-Cauchy equation) is x2y" + Ixy! + 9y=0 The Wronskian is . nM 7 xO inx nov Thus, the given solutions are linearly independent. 3x77Inx +x77 + 3x77 Inx=x77 40 Woy xt ay nx EXAMPLE 43 Find a homogeneous linear ODE of second order for which the functions e-2*coswx and e~2* sin x are solutions. Also show linear independence by considering their Wronskian. SOLUTION To the given pair e~?*cosax, e7?* sinwx of solutions, there corresponds the characteristic equation (A+2-@) A424 0) =(A+2P +0 22440 4+44+07=0 Hence, the ODE is yt +4y' + (44 02)y=0 The Wronskian is Hn de &* coswx e** sinox no WO. ¥2)= te 2x A x e**(2cosmx -asinwx) e~**(-2sinox + oc =e #0 (assuming @ #0) Thus, the given solutions are linearly independent, Scanned with CamScanner second-Order Linear ODEs A General Solution of (1) Includes All Solutions I Recall from Section 2.1, that a general solution of the ODE y" + p@)y' + q@)y=0 onan open interval /is a solution of the form O YRC + V2 where J,’ form a basis of solutions of the ODE (1), and ¢, and c, are arbitrary constants. ‘The next theorem shows that a general solution to the ODE (1) always exists. 4 THEOREM 2.7 If p(x) and q(x) are continuous on an open interval J, the ODE (1) has a general solution on J. proof Let x, be any arbitrary but fixed point in J. By Theorem 2.4, the ODE (1) has a solution y,(2) on [satisfying the initial conditions VO)=1, y'@)=0 and solution y,(x) on J satisfying the initial conditions V(X) =O, "OH Q)=I Their Wronskian at x =, is AG) ¥2(%0) (x0) ¥2(%0) a 0 i[-1+0 Wp I) Co Hence, by Corollary 2.6, these solutions are linearly independent on J. Consequently, they form a basis of solutions of (1) on J, and y=c,y; + © is a general solution of (1) on J. This proves the existence of a general solution. swing that from a general solution of (1) every solution of (1) can alues of the arbitrary constants. Hence ODE (1) has no singular nnot be obtained from a general solution. We conclude this section by sho be obtained by choosing suitable v solutions, that is, solutions that cai mogeneous Linear ODE THEOREM 2.8 A General Solution of Second-order Ho1 Includes All Solutions open interval J, then every solution If the coefficients p(x) and g(x) an continuous on some y= ¥(x) of (1) on J is of the form ¥(x) = C,y@) + C220) n and C,, Cy are suitable con stants. | where y,, y, i8 a basis of solutions of (1) 01 7, the ODE (1) has a general solution wa) Proof Let y= ¥(x) be any solution of (1) on 1. By Theorem 2. p(x) = e408) + e¥2@) OM I Scanned with CamScanner 2.32 Differentiat Feuer, in We have to find suitable values ofc, c, such thaty(2)= ¥(x) on. We choose any fixed, show that we can find constants c,, c, such that y and i derivative y’ agree with Yant eg derivative ¥' at xp, That is, y(%9) = ¥(%q) and y""(%q) = (¥q). Expressed in terms of (1), We a Ve ©1180) + &2 ¥2(%0) = mf e191 (9) + 62,9380) = Y'@0) +2) This isa linear system of 2 equations in two unknowns cy, ¢. This can be expressed in may, 10) eo ¢ [re] yiG0) y2Go)jle2] LY"Co). rm as The determinant of the coefficient matrix is HC) ¥2(%0) ¥, C0) y2(%0) = WO ¥2) Go) Since y,, y» form a basis, they arc linearly independent and hence their Wronskian is not zero by Corollary 2.6. Thus, the system (2) has a unique solution c, = C,, cy = C,. Substituting these values in (1), we obtain the particular solution Y*G) = Cy) + Cy) Now since C,, C; is a solution of (2), we have Y*G) = Fy yy) = ¥'Crp) | This means that y* and Y satisfy the same initial conditions at x. The uniqueness theorem (Theorem 2.4) now implies that y* = ¥ on J, that is, y* and Y are equal everywhere on J. This completes the proof. | ; eee EXERCISE2@ Find a second-order homogeneous linear ODE for which the given functions are solutions. Also show linear independence by considering their Wronskian. 1. 605%, e358 2, e0St, o-05s 3.22, x3 4. e348, e254 5.2, xe? 6. x5, 05 7. x-!cos(Inx), x~'sin(In.x) 8 e-*cos0.8x, e-*sin0.8x, 9.e% xe 10, +925, 5925 Inx 1. y"+3y'- L75y=0, W=-4e* 3, x2y" + 2xy’— 6y=0, W=—Sx? Scanned with CamScanner seen rire ODEs a3 4y=0, W=e-* 6 xy"—075y=0, W=-2 +2y=0, W= 8. yey" Léty= 0, W=08e-2 10. xy" +0.5xy' +0.0625y=0, W= x75 NONHOMOGENEOUS ODEs section we consider the general second-order nonhomogeneous linear ODEs y" + p@)y' + q@)y =r) (1) sere p(2), 9@), and r(x) are continuous on an open interval J, and r(x) # 0 on J. We shall see tat there is 2 close connection between solutions of (1) and solutions of the corresponding homogeneous ODE y" + p@)y' + q@)y =0 2) We begin by introducing the concepts of “general solution of (1)” and “particular solution of (1)” which are defined as follows. 2.6 In this DEFINITION General Solution, Particular Solution ‘A general solution of the nonhomogeneous ODE (1) on an open interval / is a solution of the form ye) = y,@) + ¥,@) w(3) where y, = ¢,y, + ¢,) is a general solution of the corresponding homogeneous ODE (2) on Tand y, is any solution of (1) on J containing no arbitrary constants. Aparticular solution of (1) on / is a solution obtained from (3) by assigning specific values to the arbitrary constants c, amd c, in yy, The following result shows that a general solution as defined above actually satisfies (1) and thus justifies the above terminologies. THEOREM 2.9 Relations of Solutions of (1) to Those of 2) (0) The sum of a solution y of (1) on some open interval J and a solution 7 of @) on Lis a solution of (1) on I. In particular, (3) is a solution of (1) on J ee (®) The difference of two solutions of (1) on Js a solution of (2) on Te Proof (a) Let L[y] denote the left side of (1). That is, Ly] =y" + P@y + a@y Then for any solutions y of (1) and ¥ of (2) on I, we have Ly] =r@ and LL¥]=0 (4) Consider Ly + 5] =O + 3)" + POO* By +q@ytH) =o" + 5") POY! + FD FIO * %”y Scanned with CamScanner Differential Equa mn =(y" + py! + goody) + "+ psy" + g(x) ¥) = Ly] +09) =r(x) + 0= r(x) (using (4) This proves that y + J is a solution of (1) on J. (b) Let and y* be any two solutions of (1) on J. Then Ly] =r@) and L[y*]=re) (5) Consider Ly =") = 0 -y9)" + POO =") + AED = =(" —y*") + pO" -¥*) + OY = ("+ ply’ + gly)" * peo" + aed") = Ly]-L"] =r(x)—r(x)=0 (using (5)) This proves that y— y* is a solution of (2) on J. s. The (), We saw in Section 2.5 that a general solution of homogeneous ODE (2) includes all solutio following theorem (proof omitted) states that the same is truc for nonhomogencous O1 THEOREM 2.10 A General Solution of a Nonhomogeneous ODE (1) Includes All Solutions If the coefficients p(x), q(x), and the function r(x) in (1) are continuous on some open interval J, then every solution of (1) on /is obtained by assigning suitable values to the arbitrary constants ¢; and cy ina general solution (3) of (1) on J. 2.7. FINDING A PARTICULAR SOLUTION OF THE NONHOMOGENEOUS ODE Consider the second-order nonhomogeneous ODE "+ p@)y' + a@)y =r) w(1) and the corresponding homogeneous ODE y" + py’ + q@)y =0 (2) We know that a general solution of (1) is of the form YrANtY, y is a general solution of the homogeneous ODE (2) on J and y, isa particust solution of (1). Thus, in order to solve the nonhomogeneous ODE (1) or an initial value problem for (1), we need to solve the corresponding homogeneous ODE (2) and find any particular solution y, of (I) The question is : How can we find a particular solution y, of (1) ? Scanned with CamScanner seeond-Onder Linear ODEs 235 sere are two methods for finding a particular solution. (a) Method of undetermined coefficients. (i) Method of variation of parameters. she method of undetermined coefficients is straightforward but works only for linear ODEs of the form y" tay’ + by =r(x) whore the coefficients a and b are constants and the forcing function r(x) belongs to a restricted Sass of functions. The method of variation of parameters is more general and works for every function r(x). 2.7.1 Method of Undetermined Coefficients In this section we shall discuss the method of undetermined coefficients that is used to find a particular solution y, of nonhomogeneous linear ODEs of the form y" + ay! + by = r(x) =) where the coefficients a and b are constants and the forcing function r(x) is an exponential sine or cosine, or sums or products of such functions. Note that all these functions have deri milar to r(x). Thus, it is reasonable to choose a form for y, similar to r(x), but with unknown coefficients to be determined by substituting that y,, and its derivatives into the ODE. For example, if r(x) is of the form ae“*, where a and @. are constants, then the choice for y,, would be y,, = Ae®*. Similarly, if r(x) is either acosfx or asinBx, then, because of the rules fr differentiating the sine and cosine functions, the choice for y,, would be J, = Acosx + BsinBx functioi, a polynomial function, The choice ofy, for each of the important forms of r(x) is shown in Table 2.1. TABLE 2.1 : Method of Undetermined Coefficients Form of r(x) Choice for y,(x) Constant: @ Constant : A axtb Ax+B ax" Appts Apc 89S haath Ay ae** Aet* acosBx or asinBx AcosBx + BsinBx e** [AcosBx + BsinBx] ae“*cosBx or ae sinBx (4,x" +4, yx"-1 +... + Age™ (4,x"+4,_ 1x") +... + Ap)[AcosBx + Bsin Bx] x14... + Ae® [AcosBx + BsinBx] ere x" cosBx or ax" sin Bx | 2x"e**cosBxorax"e™*sinBx | (4,x"+A ne corresponding rules for choosing the form of the particular solution y, are described as ‘ollows, Scanned with CamScanner Differential Equation, AN6 ‘ Ee ‘1 (x) iv wainglo torm, f(x) It one of the function 1 aslo Rule ‘Tho FAP een nent ing and determine ie undsteming nn in Tblo 2.1, we choone y), nd darting it TY iting», and In derivatives into the given differential equation, eooflcionts by substituting y, Pa tor shoice for y, happens to b Modification Rule “hate ape i term in on che : ee hi 6 8 HOtion Af tho corresponding homogenous ODE and therefore Toei tion of the nonhomogeneous ODE, In auch a cave, we multiply our choice of y, by x(or by x? if ig solution corresponds to 4 double root of the characteristic equnliot ot the correspondin i corre! ling, ho ce! homogeneous ODE) so that no term iny, inn olution of the corresponding homogeneous ODE, Sum Rule ‘This rule iv applied when r(x) is a sum of functions in the first column Of Table dal. In this ease, the choice for y, Is the sum of the functions in the corresponding lines op the second column of Table 2.1, a Remark In practice it is a good iden to solve the corresponding homogencous ODE before making tho choice for the particular solution Yp of the nonhomogencous ODE, ‘The above rules are best illustrated with the help of following example, EXAMPLE 44 Determine a suitable choice for Y, for each of the following nonhomogeneous ODEs. (a) yh ty! 2pm x? (b) y"=2y" + 5y= 603" (c)_ylt= Sy! + 6p de" (da) y"*y=sinx (ce) y= dy" + 13y =e cos 3x ) yay! + dy = et SOLUTION (a) The characteristic equation of the homogencous ODE y" + y’ ~ 2y = 0 is M4rn-2=(V- IA +2) =0 Thus, the characteristic equation has two distinct real roots 1 and —2, Hence the general solution of the homogencous ODE is pode Yam eyent Since r(x) = x?, a polynomial of degree 2, we choose a particular solution of the form Vy = Ax? + Bat C. () The characteristic equation of the homogeneous ODE y" - 2y! + Sy = 0 is M-2V4+5= (K-12 +4=0 => VE142I Thus, the characteristic equation has complex conjugate roots | :: 2i, Hence the general solution of the homogeneous ODE is y= e*[c,c08 2x + ¢ sin 2x] Since r(x) = 6¢%*, we choose a particular solution of the form yp Aer, (c) The characteristic equation of the homogencous ODE y" ~ 5. y' + 6y = Ois -524+6=(A-2)(0-3)=0 Scanned with CamScanner socont-Order Linear ODEs ys the characteristic equation has two distinct real roots 2 and 3. Thus, mogencous ODE is i y= Ce" + c,e3* 2.37 the general solution of the ho = 4e*, the normal choice f Id be Ae?*, But thi ice r(x) = 4e°*, the for y, would be Ae?*. But this won't x i sina solution of the homogeneous ODE. Thus, Rake tees Aet* | we should multiph i ‘i sree should choose y, = Axe? iply our choice function by x. (The characteristic equation of the homogeneous ODE y" + y= 0 is W+1=0 ‘Thus, the characteristic equation has complex conjugate roots +i, so the general solution of the homogeneous ODE is ; Yy=c,cosx + cysinx Since r(x) = sin x, the normal choice for Y, Would be Acosx + Bos x. But this won’t work because it is already a solution of the homogeneous ODE. Thus, we should multiply our choice function by x. That is, we should choose p= [Acosx + Beos.x]. (©) The characteristic equation of the homogeneous ODE y" ~ 4y’ + 13y=0is 2-40 +13=(A-274+9=0 => 2=2437 Thus, the characteristic equation has complex conjugate roots 2 + 3, so the general solution of the homogeneous ODE is y, = e2* [c,c0s 3x + ¢,sin 3x] Because r(x) = e?* cos 3x, the normal choice for y,, would bé e*[c,cos 3x + c,sin 3.x]. But this choice won’t work because it is already a solution of the homogeneous ODE. Thus, we should multiply our choice function by x. Thus, instead, we should choose yp = xe** [c,cos 3x + ¢,sin 3x]. () The characteristic equation of the homogeneous ODE y" ~ 4y' + 4y=0 is 42-4044 =(A-2)7=0 Thus, the characteristic equation has the real double root 2, so the general solution of the homogeneous ODE is yy = (c; + exde* Because r(x) = e2*, the normal choice for y, should be 5,=Ae 7 But this choice won't work because it is already a solution of the homogeneous ODE. Thus, we : i = Axe”*, But this won't should multiply our choice function by x. That is, we should choose y, = Axe?*. But Work either, y, = Axe2* is also a solution of the corresponding homogeneous ODE. So, we have to multiply our choice function by x?. That is, we choose Y= Axte2, Scanned with CamScanner Differential quar lo me 2:38 eoycibieaeaen EXAMPLE 45. Find the general solution of the equation y” ~ 2" ~ 3” x. SOLUTION Step 1. General solution of the homogeneous differential equation, ‘The first step is to find the general solution ‘y,’ of the corresponding homogeneous Opg . -2y"-3y =0 / The characteristic equation of the homogeneous ODE is -M-2-3=0 or = (AF 1)(A-3)=0 Thus, the characteristic equation has two distinct real roots 2, =~1 and A, solution of the corresponding homogeneous ODE is Vy = Ce * + Cye™. =3.Hence the geen | Step 2. Particular solution of the nonhomogeneous equation. The second step is to find ty particular solution of the nonhomogeneous ODE : yl" = 2y! — By = 2sinx | Because r(x) = 2sinx, we choose a trial solution of the type y, =Acosx + Bsinx (A, B are undetermined coefficients) > yl, = -Asinx + Beosx and yt =—dcosx — Bsinx Substitution into the original differential equation yields (Acosx ~ Bsinx) — 2(-Asinx + Beosx) ~ 3(Acosx + Bsinx) =2sinx (-44 - 2B)cosx + (24 ~ 4B)sinx = 2sinx Equating coefficients of like terms, we obtain -44-2B=0 and 24-48 =2 Solving these equations for A and B, we obtain A = 1/5 and B= 2/5. Therefore, = teosx - 2sinx ~~ 3 Step 3 General solution of the nonhomogeneous equation. The lution of the gives nonhomogeneous ODE is * * Seed soln oe | a rel Qu YAY, ty,= e+ ce + soos ~ sin. EXAMPLE 46 Solve the initial value problem y"+y= 0.0017, ¥@)=0, y(O)=15 SonOTIOR Sp 1. General solution of the homogeneous differential equation. eit Stp is to find the general solution ‘y,” of the corresponding homogeneous ODE: yrty=0 The characteristic “quation of this homogeneous ODE is 22 + 1 = 0, It has conjugate complet roots +i. Thus, the homogeneous equation has the general solution : YX) = c,Cosx + cysinx Scanned with CamScanner Onder Linear ODES. 2.39 2. Particular solution of the nonhomogeneous equation. The second step is to find th : a eular solution of the nonhomogeneous ODE ; y+ y= 0.001x2 ince 10) = 0.00185, a polynomial of degree 2, we choose a trial solution of the type Np SAN? + BYE C = Xp =ldxy+B ani yy a 2d substitution into the original differential equation yields 24+ (Ax? + Be + ©) = 0,001.2 oe Ax? + Bx + (24 +) = 0.00142 Equating the coefficients of x*, x and the constant term on both sides, we have 4 =0001, B=0 and 24+C=0 > . C =-24 =-0.002 Thus, the particular solution of the given nonhomogeneous ODE is Yp = 0.001x? - 0,002 Sep 3 General solution of the nonhomogeneous equation. The general solution of the given nonhomogeneous ODE is Y =Y, FY, = ¢c0sx + epsinx + 0.0013? ~ 0.002 (1) We now apply the initial conditions, The first initial condition yields 0) =c,-0.002=0 = ¢,=0.002 > ¥ = 0.002 cosx + c,sinx + 0.001x? - 0,002 > y! =-0.002sinx + c,cosx + 0.002x The second initial condition yields yO) =e, = 1.5 Thus, the particular solution of the initial value problem is y = 0.002sinx + 1.5cosx + 0,001x? - 0,002. EXAMPLE 47 Solve the initial value problem yl + By! + 2.25y =—10e!5*, (0) vO) =0. [Delhi Univ, GE-3, 2017] SOLUTION Step 1. General solution of the homogeneous differential equation. The first step is to find the general solution ‘y,’ of the corresponding homogeneous ODE : yl + By! +.2.25y = 0 n of this homogeneous ODE is sh ( . _The characteristic equatio Scanned with CamScanner Differential Equayg ms I double root 4 = —1.5. Hence the corresponding homogencous ODE has the Bene) 2.40 It has the real solution re Mey Feet ™ nonkomogencous equation. The second step is to fing a Particular solution of the Step 2. r homiogencous ODE : particular solution of the non! : yi + By! + 2.25y = -10e"! -18¢, the normal choice for y, would be Ae, But this won't wo, Because r(x) = -10xe : because 4e-5* ig a solution of the corresponding homogeneous ODE. Thus, we should muti, our choice function by x. That is, we should choose y, = Axe'*, But this won’t work either, y = Axe75* jg also a solution of the corresponding homogeneous ODE. So, we have to multiply, our choice function by x?, That is, we choose yt AxteV5 Calculating the derivatives of y,, we have = Yh, = AL-L.Sx2e 5 + verhS4] = AQ ~ 1.527)“ and yf =AL-LS(2x— LSx)e15* + (2 — 3u)e"!5*] = A(2 — 6x + 2.25x2)or18 ‘Substituting these expressions into the given differential equation yields A(2 — 6x + 2.25x)e“15* + 3A(2x — 1.5x2)e7}* + 2.25Ax2e715* = —10 e715 Cancelling the factor e~!5*, we get A(2 — 6x + 2.25x?) + 3A(2x — 1.5x2) + 2.25Ax Comparing the constants term, we obtain 2A =-10 > A=-5 Thus, a particular solution of the given nonhomogeneous ODE is Jy = ~Sxte“l5e Step 3 General solution of the nonhomogeneous equation. The general solution of the given nonhomogeneous ODE is Y =Iy ty = (Cy + egret" — Sx2enlse wal) Step 4 Solution of the initial value problem. We now use the initial conditions to determine constants c, and ¢,, The first initial condition yields ' 9@) =c,=1 Differentiation of y yields ¥" (e150, ~ 1.Seqx)e!5* — 10xe“!S¥ + 7.5.22 15" Setting x = 0 and using the second initial condition gives Substituting th YO =eq-15=0 => = 15, =1.5 ig the values of c, and c, in (1), the particular solution of the initial value problem Scanned with CamScanner sseoond-Order Linear ODEs e pXAMPLE 48 Solve the differential equation ; yn 241 ~ Sy 6y=4e2, soLUTION Step 1. General solution of the homogeneous di | , ' ferential ‘he first step isto find the general solution ‘y,’ of the corresponding ae | y"~Sy'+ 6y=0 geneous ODE : | ‘The characteristic equation is W-S+G=0 or (L-2K-3)=0 Thus, the characteristic equation has two distinct real roots 2 homogeneous ODE has the general solution => =2 or 4=3 1 = 3and 2, = 3. Hence the Yy= Ce + c,e* Step 2» Particular solution of the nonhomogeneous equation. The id i garticular solution of the nonhomogeneous ODE : ieee coer yl" — Sy! + 6y = de Because r(x) = 4e2%, the normal choice for y,, would be Ae?*. But this choice won't work because this function is already a solution of the corresponding homogeneous ODE. So we have to multiply our choice of y, by x. That is, we choose Jy = Axe™ Calculating the derivatives of y,, we have yi, =2dxe*+Ae™ and y= 4Axe™* + de® Substituting these expressions into the given ODE yields (44xe?* + 4402") — 5(2Axe* + Ae**) + 6(Axe?*) = 4e* > —Ae™* =4e* > A=-4 Thus, y= —4xe? is a particular solution of the given nonhomogeneous ODE. Step 3 General solution of the nonhomogeneous equation. The general solution of the given nonhomogeneous differential ODE is y =I ty, = Oem + cye** — Axe™. EXAl Solve the initial value problem ul ) 5 yi ty Dy = 4c, y(0)=7, y'(0)=0. SOLUTION Step 1. General solution of the homogeneous differential equation The frst step is to find the general solution ‘y,’ of the corresponding homogencous ODE: yt ty'- 12y=0 The characteristic equation is p2+2-12=0 or (-3)R4A)=O wo distinct real roots 2, = 3 and = 7-3 01 d=-4 = -4, Hence the Thus, the characteristic equation has t h homogeneous ODE has the general solution . F yy, =ee* tee Scanned with CamScanner 2.42 Differential Equatio ms Step 2. Particular solution of the nonhomogeneous equation, The second Step is to fing particular solution of the nonhomogeneous ODE : a yl ty! Ly = 40% Because 7(x) = 4e?%, we choose a trial solution of the type y, = de® (A isan undetermind coeffici Y% "= dder > yya2de™, y= Ade Substituting these expressions into the given ODE, we obtain 4Ae?* + 240? — 12de** = de® > 44+24-124 =4 (cancelling ¢?*) => A=-2/3 Thus, the particular solution of the given nonhomogencous ODE is 2 ax aac Step 3 General solution of the nonhomogeneous equation. The general solution of the given nonhomogeneous ODE is s x 2 2x Y AWA Yp= ge + ce — 3° af) Step 4 Solution of the initial value problem. We now use the initial conditions to determine constants c, and c,. Setting x = 0 and using the first initial condition gives ¥(0) =¢, +e, -23=7 wal) Differentiation of y yields ¥ aoe age Se From this and the second initial condition, we have 4 (0) = 34 ~ dey - 5 = 0 a) Solving Eq.(/) and (ii) simultaneously, we find that c, = 32/7 and c, = 65/21. Substituting these values of c, and c, in (1), the particular solution of the initial value problem is = ry Spy ore a pee et Set EXAMPLE 50 Solve the initial value problem y" ~3y' + 2y = 3e* — 10cos3x, YO)=1, yO) =2. SOLUTION Step 1. General solution of the homogeneous di The fi FB lifferential equation. Hiss sep iso find the general solution ‘y,’ of the corresponding homogeneous ODE : y" By! + 2y=0 Scanned with CamScanner syomt-onter Linear ODEs aa hi characteristic equation The 5 3A4+2=Q-1)A-2=0 vo distinct eal roots = I and A= 2, so the general solution of the homogeneous ODE is ie Jy = Cye* + eye 2, Particular solution of the nonhomogeneous equation. The second step is to find a up 2. ait solution of the nonhomogeneous ODE : pa y — 3y' + 2y = 3e-* - 10cos3x Because 1@) = 3e°* ~ 10cos3x is @ sum of two functions in the first column of Table 2.1, we choose for J, the sum of the functions in the corresponding lines of the second column, That is, we choose Jp = Ae + Boos 3x + Csin3x > JY, =~Ae~* — 3Bsin3x + 3Ccos3x 5 Y}, = Ae — 9Bcos3x — 9Csin 3x Substituting these expressions into the given differential equation, we obtain (e™~ 9Beos3x — 9Csin 3x) — 3(-Ae-* — 3B sin3x + 3Ccos3x) + 2(de~™ + Boos3x + Csin3x) = 3e-* — 10cos 3x Regrouping the terms yields Sde~* + (-1B - 710) cos3x + (9B ~ 71C)sin3x= 3e-* — 10 cos 3x Equating the coefficients of e~*, cos3x and sin3x, we obtain 6A =3 -7B-9C =-10 9B-7C=0 Solving these equations, we get A = 1/2, B= 7/13 and C = 9/13. This gives the particular solution Step 3 General solution of the nonhomogeneous equation. The general solution of the given Nonhomogeneous ODE is 1, 7 9. yay, ty, = e+ oe + oe + Tyoos3e + imax ay Seo 4 Solution of the initial value problem. We now use the initial conditions to determine Constants ¢, and c,. We need to derivative px_ lx 2, ZZ eos3x v= ge'+ dee het Beings + 2 Scanned with CamScanner 244 Using the initial. conditions, we have 1 =1 1 = 0) 44643 t 5 1,2. 0) = 4 420-915 1. = © substituting these val Solving these equations, we obtain ¢ =~ 3 > 73° a stituting alues of c, ang eh a Eq.(1), we obtain the particular solution of the given initial value problem | 9 Lig Gop tets Leostx + Ssin3x, chet Sepn gee pee gti y mined coefficients to find the particular solution g “ Use the method of undeterr f EXAMPLE 51. Use th find its general solution. [Delhi Univ. GE.3, am ODE y" +4y! + Sy= 25x? + 13sin2x. Also SOLUTION Step 1. General solution of the homogeneous differential equation, The first step is to find the general solution ‘y, of the corresponding homogeneous ODE - "+ dy! + 5y =0 The characteristic equation is 4244245=0 or (A+2)2+1=0 = AR=-24; Thus, the characteristic equation has complex conjugate roots ~2 + i. Hence the comresponcig homogeneous ODE has the general solution é yy, =e [c, cose + ¢,sinx] Sep 2. Particular solution of the nonhomogeneous equation. The second step is to fini particular solution of the nonhomogeneous ODE : yp" + Ay! + Sy = 25x2 + 13 sin2x | Because the forcing function r(x) =25x? + 13sin2x is a sum of two functions in the first column of Table 2.1, we choose for y, the sum of the functions in the corresponding lines of the second column. That is, we choose Yp = Ax? + Bx + C+ Deos2x + Esin2x (1) This implies yy, = 2dx +B -2Dsin2x © 2Ecos2x yt =2A ~ 4D cos2x— 4B sin2x Substituting these expressions into the given differential equation, we obtain (24 ~ 4Dcos2x ~ 4E sin2x) + 4(2Ax + B - 2Dsin2x + 2Ecos2x) + 5(Ax? + Bx + C+ Deos2x + Esin2x) = 25x? + 13sin2x Regrouping terms, we get (24 + 4B + SC) + (8A + SByx + 5Ax? + (8E + D)cos2x + (E — 8D)sin2x = 25x? + asin Scanned with CamScanner vorter Linear ODEs cont : efficients of like powers of x, and th i tee ating © ;, and the coefficients of cos? i : fav" yA + 4B + SC= 0, 8A+5B=0, SA=25, 8E+D=0, eee sig equations for these 5 unknowns, we obtain = cl Aad 8, C=44, D=-16, £=0.2 seit ues of these unknown in Eq, (1 i i ppstituting the val iq. (1), we obtain a particul: i ' a rpomogeneoUS ODE Particular solution of the given Jp = 5x? 8x +44 ~ 1.6c0s2x + 0.2sin2x 3, General solution of the nonhomogeneous equation. The general solution of the given ronkomogencous ODE is J. = Yq typ =v? Le cosx + c,Sinx] + 5x?~Bx + 4.4 ~ 1.6cos2x+0.2sin2x EXAMPLE 52 Solve the initial value problem yl + dy" + Sy =e + 40cos 10x 190sin 10x, y(0)= 0.16, (0) = 40.08. SOLUTION Step 1. General solution of the homogeneous differential equation. ‘The first step is to find the general solution ‘y,’ of the corresponding homogeneous ODE : y" + dy! + Sy =0 ‘The characteristic equation is 22+ 2K45=0 or Qt)?t+4=0 = A+1L=+£2i or Hence the corresponding homogeneous ODE has the general solution y, = e*[e,c0s2x + ¢,sin2x] Step 2. Particular solution of the nonhomogeneous equation, The second step is to find a particular solution of the nonhomogeneous ODE : : y+ Dy! + Sy = e05* + 40.cos 10x - 190sin 10x Because the forcing function r(x) = €95* + 40c0s 10x - 190sin 10x . is a sum of two functions in the first column of Table 2.1, we choose for y, the sum of the fimetions in the eorresponding lines of the second column. That is, we choose y, = Aeds + Boos Ox + Csin 10x (1) Then yf, = 0.54205" — L0Bsin 10x + 10Cc0s 10x > yt = 0.25Ae°% - 100B cos 10x - 100C sin 10x Substituting these expressions into the given differential equation, we obtain (0254605 190Be0s 10x - 100Csin 10x) + 2(0.5de®S* - 10Bsin 10x + 10Ce0s 10x) 4 5(de®S* + Boos 10x + Csin 10x) = e%S* + 40c08 10x 190sin 10x Scanned with CamScanner 2.46 Pifferentiat Equation Regrouping terms, we get os osx «+ (-20B — 95C)sin 10x = €°5* + 40 cos 10x ~ 199, 6.254005" + (-95B + 20C)cos 10x + (-20B — 950) 90 sini, Equating the coefficients of e***, cos 10x and sin 10x, we obtain | 6.254 =1 | —95B + 20C = 40 | —20B -95C =-190 Equation (i) gives'A = 1/6.25 = 0.16. Solving Eqs.(éi) and (ii!) simultaneously for B and ¢, y. | obtain B = 0 and C = 2. Substituting the values of 4, B and C in Eq.(1), we obtain a partieyi, | solution of the given nonhomogeneous ODE | Jp = 0.16e%5* + 2sin 10x | Step 3 General solution of the nonhomogeneous equation. The general solution of the given nonhomogeneous ODE is | y =y, typ = EME, c082x + c,sin2x) + 0.162% + 2sin 10x (2) Step 4 Solution of the initial value problem. We now use the initial conditions to determing constants ¢, and c,. We need the derivative | y! =e7*(-2e, sin2x + 2c,c0s2x — ¢,cos2x- c,sin2x) + 0.086 + 20coslr Using the initial conditions, we have y(0) = 0, + 0.16 = 0.16 | y'(0) =-c; + 2c, + 0.08 + 20 = 40.08 | Solving these equations, we obtain c, = 0, c, = 10. Substituting these values of c, and c, into Eq.(2), we obtain the particular solution of the given initial value problem y =10e-*sin2x + 0.16e%5* + 2sin 10x. | EXAMPLE 53 Solve the initial value problem: y" + 3y’ + 2y=22. | SOLUTION Step 1. General solution of the corresponding homogeneous differential equation. The characteristic equation of the corresponding homogencous differential equation is W+3K4+2=0 ie, (A+NDQA+2=0 => A=-1,-2 | > a =-1, -2 Distinct real roots | Thus, the characteristic equation has two distinct real roots -1 and —2. Hence the homogenco¥s | ODE has the general solution Dy = Cet + Ce ‘Step 2. A particular solution of the nonhomogeneous differential equation. : ‘We now obtain a particular solution, Nps of the nonhomogeneous ODE yt + By! + 2p =x? Scanned with CamScanner order Linear ODEs we 2, a polynomial of degree 2, we choose for y, " ayer, ert yp = AX? + Bx +C a trial soluti t a solution of the form: Vy =2dAx+Bo and y"=24 se expressions into Eq.(1) yields 2A + 3(2Ax + B) + 2(Ax? + Bx + C) =x? 2Ax? + (6A + 2B)x + (24 +3B + 20) =x? sting the coefficients of x?, x, and the constant term on both sides, we have sosiing thes 24=1 wi) 64 + 2B =0 wii) 24 +3B+2C =0 Git) a(p yields A = 1/2. Substituting this value in (i), we get 3+2B =0 => B=-3/2 substituting the values of A and B in (iii), we get 1- ; +2C =0 > c= t Hence a particular solution of the given nonhomogeneous ODE is given by 1. 3,7. Y= ae 5ktG Step 3 General solution of the nonhomogeneous equation. The general solution of the given nonhomogeneous differential equation is te eety te 3,47 yeytys ett gets ote dead _ EXERCISE 2.5 Use the method of undetermined coefficients to find a particular solution of each of the following differential equations. Also give a general solution. 1 y"-y'-6y = e2* 2. y"—4y' + 3y = 3x? - dy" tayeex | 4. y"+y'-2y=sinx Sy" + y= sinx : 6. y"—4y' + 3y=e* Tey" +3y"— Oy = 3x2 8. y"ty!-2y=x? 5. y"+ dy" + 3.75y = 109 cos Sx 10. y"— l6y = 19.2e%* + 60e* TL y"4dyt 4 4y= ent sin 2x 12. yp" + 6y"+ 73y = 80e* cos4x Bey" + Lady = 24cos1.2x ; 14. y" + 9y = 18x + 36sin3x 18. yr —syy + 6y=4e* 16. y" —4y =xe* + cos2x Scanned with CamScanner 2.48 Differential Faqation, Use the method of undetermined coefficients to find the solution of each of the following ing value problems. 17. y" —4y' + 3y = 5Se~4*, (0) = 4, (0) 18, y" + 4y = L6cos2x, (0) =0, y'(0)= 19. y" — By" + 2.25y = 27(x2— x), (0) = 20, (0) = 30 20. y" + 1.2y! + 0.26y = 1.225%, y(0) = 3.5, y'(0) = 0.35 21. y"— 2p" = 12e* — 8e-, y(0) = -2, y"(0) = 12. 1 — > ANSWERS” ; 8 ou 1 2x = cet 34 2,8, Le y= qe + ee — Fe 2, ya et Gent + Sree x -2r_ 1 - 3. p= ecos2x + cp sin2x+ = 4, y= ee" + ce ~ TF (cosx+3sin2) x 3x, 1: S.y= 6,608 + epsinx = 3 xcoss 6. y= qe" + ce 4 Sxe sx 32 9 57 -2x 12 1 3 Ly = qe* + ee apt ta 500 8. y= ge* + oe ng age 9. y = c,e-15* + c,e-25* — 2.72 c085x + 2.56 sin 5x 10. y= ce + c,e-** — 2.4xe"* — de I. y= (q tex)e-* - j 28 sin 2x 12. y = e~**(Acos8x + Bsin8x) + e* (costx + jens] 13. y = ¢, cos 1.2x + c,sin1.2x + 10xsin 12x 14. y = Acos3x + Bsin3x + 2x—6xeos3x 18. y= cye?* + ce Are? dx -2x_(1 2), 1 * 16. y = ge + e* - |x = }e— Leggo y= oy 2 (3 2) 3 x 1S jae, Lae . Cia ea ol 18. y = 4xsin2x - 0.5% 19. y= AlCl + a)elSr 3x2 + 5x4 4] 20. y = e511 5 ¢05 0.5.x — sin 0.53] +2¢ 21, y =2e?*-3 + 6xe2— e-2x, | 2.7.2. Method of Variation of Parameters In this section we shall discuss yet another method for finding a particular solution of second-order nonhomogeneous linear ODE that is given in the standard form y" + p@)y' + g@e)y = rex) va) Scanned with CamScanner stoner Linear ODEs a0 2.49 re isa eneous per ps) JO) and (8) ae Continous on some o ae eannection between solutions of OE (1) and ope Din interval. We have seen that the Solutions of the corresponding homog y" + pady" + g(xy =0 in tics general solution of (1) is of the form aa YEA, here, 8 general solution of the correspondin ‘Shtion of (1). If the foreing function r(x) con sinBax, we ean find a particular solution n in Section 2. 8 homogeneous ODE (2) and y, is any particular sists of sums or products of x”, e**, cosBx or by the :d coefficients, as we have is valid only for restricted class of functions r(x), itis ible to have a method that applies to more general differential equations (1), which we shall naw develop. It is called the method of variation of parameters. The method of variation of parameters uses a pair of linerarly independent solutions of the corresponding homogeneous ODE (2) to construct a particular solution of ODE (1). sin 3p by the method of undetermine Since this metho We shall develop this method in connection with the general second-order linear ODE (1). Suppose swe know that y,(x) and y,(x) are linearly independent solutions of the corresponding homogeneous ODE (2). A general solution of the corresponding homogeneous equation is therefore given by Yi@) = CV) + ey) The idea in the method of variation of parameters is to replace the arbitrary constants (“the parameters”) c, and c, by functions u(x) and v(x) so that the resulting function YX) = UG) y¥{@) + V@)y,@) +(3) isa particular solution of the nonhomogeneous ODE (1) (this suggests the name of the method, variation of parameters). : The functions u(x) and v(x) will be governed by a system of two equations, one of which is derived by requiring that Eq. 1) is satisfied, and the other of which is chosen to simplify the resulting system. Differentiating (3), we obtain Ale) = WED) + VEIVIG) + WEI) + YOO) A) We simplify y'(x) by imposing the condition u'@)y,@) + V@)y8) = 0 (5) With this condition imposed, Eq(4) reduces to a hla) = uC) + POO i. Differentiating (6), we obtain a wy'e) o VM) = ula) fla) + VOD PIC) # HONE * vey a t Substituting y, and its derivatives according to (3), (6), (7) into (), wees i 9y400] , i jan Cx) yS(a)] + RODEO) + VODA Leyte) + vy) + WOOYIA) + OY = gee 0) * vG)9,@)] = 7@) Scanned with CamScanner 2.50 Differential Eauayo ny Regrouping the terms, we get UO)LYTOD + PESO) + GOIN AD] + VOEILZE) + PEO) + g(x)y,0)] + W@Oy{@) + VOI Since y, and yy are solutions of the corresponding homogeneous ODE (2), the expressions in je s in (8) are identically zero, Eq.(8) thus reduces to first two brac! 40) — YOO) nO) 20) HO) 0) Since y, and y, are linearly independent solutions of the corresponding homogeneous ODE, we know that Wo yy4 0 Hence the system (10) has a unique solution. Using Cramer’s rule, we find that (x)r@) w@) = Integrating, we get WO) Therefore, a particular solution y,, of Eq.(1) is given by Jp = UY) + HG where u(x) and v(x) are defined by (11). We now outline the procedure for finding a particular solution of the nonhomogeneous ODE using the method of variation of parameters. | (LD) Procedure for Finding a Particular Solution of the nonhomogeneous ODE y" + p@)y' + q@y =r) Step1 Find linearly The first step is to find homogeneous ODE independent solutions of the corresponding homogeneous ODE: pair of linearly independent solutions {y(x), y4(x)} of the corresponding »" + pQ@)y' + a@)y =0 Scanned with CamScanner e eont-order Linear ODEs ep2 Fins a general olution of the corresponding homogencous 0 “ . , . : se general solution y, of the corresponding homogeneous ODE ig a ; Yn = CY, (2) + chy, (x) ie ; ’ wl 3 Find a particular solution of the nonhomogeneous ODE: ° sre particular solution y, of the nonhomogeneous ODE is given by Yp = MONO) + V(@)y,0), (x) and v(x) are given by where 1 = (22@)r) ari 00) = Fro, on v= Ting EXAMPLE 54 Use the method of variation of parameters to solve the nonhomogeneous ODE y" + By" + 2y = 300, [Delhi Univ. GE-3, 2016} SOLUTION Step1 Find linearly independent solutions of the corresponding homogeneous ODE ‘The corresponding homogeneous ODE is y" + 3y' + 2y=0 (1) The characteristic equation is W+3R42=0 or (At IA+2)=0 = A=-1,-2 Thus, the characteristic equation has two distinct real roots =1 and —2. Hence the functions y,(x) = e~* and y,(x) = e-?* are two solutions of the homogeneous ODE (1). These solutions are linearly independent because their Wronskian is HOD 2 HO) yy Wy ¥2) = Step 2 Find a general solution of the corresponding homogeneous ODE The general solution of the corresponding homogeneous ODE (1) is Vy = eye* + cne-* Step 3 Find a particular solution of the nonhomogeneous ODE By the method of variation of parameters, a particular solution of the given nonhomogeneous ODE is Yy =uee* + v(xe™* (2) Wheré u(x) and v(x) are given by =y2@)r@) WO» ¥2) H@)re) 9) = Sing * EGO) a = 30fe* = 102 u(x) = -*0e") , 2 agfste 150!" eres = -30fe Scanned with CamScanner /ifferentiah Eau atic 2.52 Substituting the expressions of the given nonhomogeneous ODE = u(xye* + vee = 10e3*- e* for u(x) and v(x) back into equation (2), We get the particular. ln, Sly n | T5e¥e* = 2 5e2x yp . the general solution of the given nonho + yy =e + OE mmogencous ODE is Hene 2x + 2,.5€%. ya 0 solve the nonhomogeneous o Dg EXAMPLE 55 Use the method of variation of parameters t yl ty = tanx. SOLUTION Step1 Find linearly independent solutions ofthe corresponding homogeneous Ong The corresponding homogeneous ODE is | aan ~(l) The characteristic equation is W+1l=0 > =H Thus, the characteristic equation has two complex conjugate roots +i. Hence the functoy +9402) =c08x and y,(x) = sin x are two solutions of the homogeneous ODE (1). These solution ze linearly independent because their Wronskian is HG) y2(3) 3@) y4(x) cosx sinx = cos2x + sin? ir xt sinx cosx| = °0S'* + sin?x=1#0 Wp Yr Step 2 Find a general solution of the corresponding homogeneous ODE The general solution of the corresponding homogeneous ODE (1) is Jy, =e, cosx + cy sinx Step 3 Find a particular solution of the nonhomogeneous ODE By the method of variation of parameters, a particular solution of the given nonhomogeneous ODEs Yp = u(x) cosx + v(x)sinx, 2) where u(x) and v(x) are given by Yo(x)r(x) in? cos’ x u(x) =| de = | - =-(88@242-(—=4 Woe) Jrsinxtanxde por dx ai = J(cosx-secx)de = sinx—In|secx+tanx| and _ fu@)r@) v(x) = Wooo* = foosxtanxde = [sinxde =—cosx Substituting the expressi i \: : ‘pressions for u(x) and v(x) back int i +t the particular: sol of the given nonhomogeneous ODE (x) into equation (2), we get the pal ¥p = cosx[sinx ~ In|secx + tanx |] — sinx cosx Scanned with CamScanner g-Order Linear ODEs 2:53 eset general solution of the given nonhomogeneous ODE is = y, ty, =e cosx + csi i Y= Yy Fp Cy C08x* cy sinx + cosx[sinx — In|seex + tan. |] ~ sin-xc = sinx cosx, (PLE 56 Use the method of variation of parameters to find a g nonhomogeneous ODE general solution of the sollowin a ; indy! ty =etsinxe [Dethi Univ. GE-3, 2018} soLUTION The general solution of the nonhomogeneous ODE oo yu = 2y' + y= exsinx a1) js of the form V=Vyt Vp» wyherey, is the general solution of the corresponding homogeneous ODE y"—2y'+y=0 (2), andy, isa particular solution of the ODE (1). Genero! solution of the homogeneous ODE (2) ‘The characteristic equation of the homogeneous ODE (2) is -2A+1=0 => (A-1P=0 => AH=L1 Thus, the characteristic equation has the real double root 4 = 1. Hence the functions jy(2) = e* and y,(2) = xe* are two solutions of the homogeneous ODE (1). These solutions are linearly independent because their Wronskian WG) v2) xe* ; Q = 2x _ yet = 2" yy) ¥2(2) (@& + Ie* — xe #0 = (x4 Ne WO) = The general solution of the homogeneous ODE (2) is y= eyet + eyxe* Particular solution of the nonhomogeneous ODE (1) By the method of variation of parameters, a particular solution of the nonhomogeneous ODE (1) is Yp = ulxe* + vex)xe* (3) where u(x) and v(x) are given by ¢ u(x) = =[ par = = -fxsinxde @)r@) = vi): So. 9) xo0sx+[ cosas | = xeosx— sinx = Jsinxde = —c0sx Substituting the expressions for u (x) and v(«) back into equation (3), we get the particular solution of the nonhomogeneous ODE (1) yp = wetcosx—etsinxxeTeOsx = —e1SD Scanned with CamScanner bis Differentic 2.54 ial Eaton Hence the general solution of the nonhomogeneous *sinx. ODE (1) is a ya yytyy sere’ tere & ape EXERCISE 2.5 i 1 Use the method of variation of parameters to find a particular solution of each of the Tolloving differential equations, Also give a general solution. 2. y"+y=tanx Ly"-y x 3. y" + 9y = 3tan3x 4 yay tye 6. y" ty =see x 8. y"—4y! + 4y =27e* 10. y"=4y' + 4y = 12e/x4 12. (D?-2D + Dy = xe" ar Tis L y= qe't ae" + axe 2. y= qcosx + cp sinx — cosxIn| secx + tanx| 3. y= ¢0083x + cp sin 3x — $eos3In|seo3x + tan3z| 4, y = qe" + cyxe* ~

You might also like