6.14. SYLLABUS 395 3.14 Syllabus partial differential equations and Complex analysis Module I partial Differential Equations (Ref. 1.- Relevant portions of sections 17.1,17.2,17.3,17.4,17.5,18.1,18.2) Partial differential equations(PDE), Formation of PDEs -elimination of arbitrary constants- elimination of arbitrary functions, solution of partial differential equation equations, Equa- tions solvable by direct integration, Linear partial differential equations of first order. - Logrange’s linear equation, Non-linear equations of the first order - Charpit’s method, Solution of equations by method of separation of variables. [8 hours] Module IT Applications of Partial Differential Equations (Ref: 1 -Relevant portions of sections 18.3,18.4,18.5) One dimensional wave equation - vibrations of a stretched string, derivation, solution of the wave equation using method of separation of variables, D’Alembert’s solution of the wave equations, One dimensional heat equation, derivation , Solution of the heat equation. (10 hours} Module III Complex Variable — Differentiation (Ref2 - Relevant portions of sections 13.3, 13.4, 17.1, 17.2 , 17.4) Complex function, limit, continuity, derivative, analytic functions, Cauchy-Riemann equations, harmonic functions, finding harmonic conjugate, Conformal mappings- map- pings w = z?, w = e*, Linear fractional transformation w = 4, fixed points, Transforma- tion w = sin z (From sections 17.1 , 17.2 and 17.4 only mappings w = z?, w =e*, w= 1, w=sinz and problems based on these transformation need to be discussed) [9 hours} Module iV Complex Variable - Integration (Ret: 2 -Relevant topics from sections 14.1, 14.2,14.3, 14.4,15.4) Complex integration, Line integrals in the complex plane, Basic properties, First eval- “ation method - indefinite integration and substitution of limit, second evaluation method ~ Use of representation of a path, Contour integrals, Cauchy integral theorem (without Proof) on simply connected domain, Cauchy iritegral theorem (without proof) on multiply (Panected domain, Cauchy Integral formula (without proof), Cauchy Integral formula for ‘etivatives of an analytic function, Taylor’s series and Maclaurin series. - [9 hours} Modules V396 CHAPTER 8. PROBLEM SOLVING SESSION Complex Variable — Residue Integration (Ref: 2 - Relevant topics from Sections 16.1, 16.2, 16.3, 16.4 ) * Haurent’s series(without proof ), zeros of analytic functions, singularities, poles, re- movable singularities, essential singularities, Residues, Cauchy Residue theorem (without proof), Evaluation of definite integral using residue theorem, Residue integration of real integrals — integrals of rational functions of cos 9, andsin 9, improper integrals of the form J2 f(z)dz with no poles on the real axis(improper integrals of the form J, f(z)dz whose integrand become infinite at a point in the interval of integration is excluded from the syllabus) Assignment : Assignment must include applications of the above theory in the concerned engineering branches References: 1. B.S. Grewal, Higher Engineering Mathematics, Khanna Publishers, 43rd Edition, 2015. 2. Erwin Kreyszig, Advanced Engineering Mathematics, 10 th Edition, John Wiley & Sons, 2016. 3. Peter V. O'Neil, Advanced Engineering Mathematics , Cengage, 7th Edition, 2012Contents 1 Partial Differential Equations 1 1.1 Formation of Partial Differential Equations ee 3 1.1.1 By eliminating the arbitrary constants Ceca as 3 1.1.2 By eliminating the arbitrary functions . . . : 10 113 By eliminating the arbitrary function ¢ from $(u, v) a eeeb 1.2. Solution of Partial Differential Equations........ 0.0... ~” 7 416 1.2.1 Solutions by direct integration . eee ee 1.2.2 Solution of first order PDE... .........«. eg 820 1.2.3 General form of First order PDE coe aeao 1.2.4 Linear and non-linear PDE... 2.0.0... eo e ad 628 1.3. Lagrange's linear partial differential equation aoe MA Mxerciseg Sih. eee eee aos 1.5 Non-Linear Equations of First Order ie #85 1.5.1 Charpit’s Method oe aia ae 16 - Exercises oy ee eat L7 Solution by Method of Separation of variables ane 47 M18 kercises' eens | 2 One Dimensional Wave Equation 55 2.1 One Dimensional Wave Equation aa 55 2.2 Solution of one dimensional wave equation... ... 87 2.2.1 Solution using method of separation of variables . 87 2.2.2 D? Alembert’s solution z eso | 2.3. Fourier Series Solution of wave equation ee tas 4 261 24 Mxercies ee Coe ee 94 3 One Dimensional Heat Equation 95 | 3.1 Introduction. ........... Sa ee ee os | 3.2 One Dimensional Heat Flow ....... See agers .. 95 3.3. Solution of Heat Equation using separation of variables... ......_ 97 3.3.1 Steady state solution of heat equation eee - 99 3.4 Fourier series solution of the heat equation ...... 0.0...” +2101 3.4.1 Both the ends are kept at zero temperature oe oe aedan 3.4.2 The left end is kept at 0°C' and right end is insulated ...... |. 11 3.4.3 The left end is insulated and right end is kept at 0°C ....... 13 | 3.4.4 When both the ends are insulated et. 1GCONTENTS 1 43. The ends are neither at zero temperatures nor insulated 1» 35 Brercise 5 a Sei . 125 4. Runctions of a Complex Variable i“ 4.1 Complex Numbers... mt 42. Polat form of a complex number +. = 43, Motliplying i - Geometrical interpretation 1 44 Extraction of roots of a complex number Fe 4.5. Neighbourhood of a point and open sets = 46 Plane Curves in complex plane. - - ie 47 Simply asd multiply connected regions «- 14 48 Bounded region _ ms 49 Exercise ee 4.10 Functions of a complex variable 1s 4411 Elementary functions of complex variable 4.111 Exponential funtion of imlexvrale 2.187 4.12 Limits and continuity 138 4.12.1 Bounded Function 138 4.12.2 Limits... 138 412.3 Continuity... 139 4.124 Properties of limits and continuity 12 4.13 Derivative of function of complex variable a2 4.18.1 Properties of derivatives : 146 4.13.2 Necessary conditions for differentiable functions 146 413.3 *Sulfcient conditions for ¢ function to be differentiable - 148 4254 Polar form of Cauchy -Remann equations 9 4.14 Analytic funetions ‘ 154 4.15 Orthogonal Trajectories” . 165 4.15.1 Singular points of analytic fanctions . 169 415.2 Harmonie function... 170 4.153. *Milne-Thomson method 175, 416 Exercise... i 4.17 Chapter Summary 178 5. Conformal Mapping 181 51 Invariants of angle made by curves 181 5.1.1 Critical points and Iavaciant points 183 52. Some elementary transformations 185 5.2.1 The transformation w=x+e, where eis a complex constaat (Trans- 522 The tas = 3¢ transformation wes where cis real constant (Magnification 523 The ranfomation w= ies e ation w=cz where cis a complex constant (Magnif- cation and Rotation) eee is 524 The transformation w = + (Inversion and reflection) 186 53. Some Important, Conformal Mappings ie CONTENTS 53.1 ‘The transformation w= 2? . 5.3.2 The transformation w = sine 53.3 The transformation w = cons 5.3.4 The transformation w = e 53.5 The transformation w = =-+2 5.3.6 Bilinear transformation (Mobits transformation)* 54. Exercise 5.5 Chapter Summary ‘Complex Integration 6.1 Definite integrals. . 62 Complex Line integral. . 6.21 The Complex Integral in Terms of Resl Integrals 6.2.2 Path independence of line integral 7 62.3 Cauchy-Goursat Theorem 6.24 Cauchy's Integral Formula 625 Cauchy's Integral theorem for Multiply Connected domain 6.3. Power series expansion of Analytic functions 63.1 Taylor series 7 : 6.3.2 Laurent’s series 64 Exercise... . 65 Chapter Summary Residue Integration 7A Singularities and Zeros TALL Zeros of Analytic Functions 7.1.2 Isolated and Non-isolated singularities 7.1.3. Different types of Isolated singularities 7.2. Residues of a function at « point 7.2.1 Methods for computing Residue 7.3. Residue Calculus - Residue Integration Method 74. Residue Integration of Real Integrals 7.4.1 Integrals of Rational Functions of cos@ and sin @ 74.2 Integrals around a semi circle 75 Exercise . 7.6 Chapter Summary Problem Solving Session 8.1 Model Question paper 82 Solution to the model question 8.3 Question paper (March 2017) 8.4 question paper (Dee 2016) 85 Question paper( April 2018) 8.6 Practice Questions 8.7 Solved question papers (S2 [Link], 2016 eee 88 Question paper (April 2018) 189 193 194 195 197 198 202 203 205, 207 207 - 2 216 air 295 226 27 231 241 241 241 23. vg 247 251 258. 265 273 275 aT 279 304 313 322 325,CONTENTS vt 8.9. Question paper (July 2017). « . ur 8.10 Quostion paper (May 2017) : a 8.11 Question paper (Angust 2016)... « om 8.12 Question paper (July 2016) eerereretrcor ae sme 813 Question paper (June 2016) 85 8.4 Syllabus Chapter 1 Partial Differential Equations Differential equations can be broudlly classified into teo types: ordinary differential equa- tions: tbat is, differential equations in which ordinary derivatives of function of one vari- ablo are present and partial differential equations: differential equations involving, pat derivatives of functions of more than one wasiable are involved A differential equation which involve partial derivatives is called partial differential equation(PDE). For the purpose of classification of partial differential equation, we define order and degree of partial differential equations similar to the one defined for ordinary differential equations. The order of the highest order partial derivative present. in the partial differential equation is called the order of the partial «lifferential equation. The power of the highest: order partial derivative present in the PDE after making each term. free of radicals and there are no terms in whiich denominator contains derivative as factors is called degree of PDB. The general form of mth order PDE involving one dependent variable 2, two independent variables x and y may be expressed us ‘one cimenstonal beat equation the two dimensional heat equation ou a ot aa’ ou, Bw ast Gi): We may oots that th above PDEs ore second oder ait dagtes PDEA. A fow of he famous PDE t btretue are Salar reltonifsion equation % oy = J(u)CHAPTER 1, PARTIAL DIFFERENTIAL EQUATIONS ca aw Korteweg-deVrie equation ou, Ou, Ou _ oa ou Ot, Pe yay 0 ot Yar * oe Euler's equation for incompressible, inviscid flow au Su. Du=-DP Mew divu =0 . Navier Stokes equation for incompresible, viscous flow Shu wu-Du=-DP a divu=0 The transport equation au ith duno Laplace equation Vvu=0 ‘The heat equation gaye , Gn Uua0 ‘The wave equation au Ge Vu=o The Poissson’s equation —Wu=f where V? is the Laplacian operator given by 2,8 @& Vesta tia Among the PDE, Laplace equation is the most important PDB. Ifu denote the cheunical Lee gee, temperature and electronic potential, then tbe Laplace equation is Fick's Han, of difusion, Fourer's law of heat conduction in steady state and Ole’ hens of leetrical conduction respectively, in its own or allied fields of science ‘We make use of the symbols p= BO: oo a: ae 0 yO 8 Boy Bae (= SS ‘where = isa function of two independent variables 2 and y et 14, FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 3 1.1 Formation of Partial Differential Equations Now we shall discuss the formation of PDE using the following, methods 1. Elimination of arbitrary constants, 2. Elimination of arbitrary functions 5, Elimination of arbitrary function ¢ from the relation ¢(u,) = 0, where u and v are functions of x,y and 2 1.1.1 _ By eliminating the arbitrary constants Here we willbe given a relation connecting one dependent variable, two or more indepen- dont variables and one or more arbitrary constants, First, we shall examine the interesting question of how they arise. Consider the equation P+Pt(z-Pad a in which the constants a and c are arbitrary. Then equation (1) represents the set of all spheres whose centers lie along the 2-axis. If we differentiate this equation partially with, respect to x, we obtain the relation Ztp2-0) =0 @) fand when we differentiate it partially with respect to y , we get y+q(z—0)=0 @) Elimuinating the arbitrary constant c from these two equations, we obtain the partial differential equation w-xqg=0 a which is of the first order. In some sense, then, the set of all spheres with centres on the axis is characterized by the partial differential equation (4). However, other geometric entities can be described by the same equation. ‘Note that in this example the number of independent variables and number of arbitrary ‘constants are equal. In general, when the number of independent variables and the number of arbitrary constants are equal, the resulting PDE will be of first order. If the number of arbitrary constants to be eliminated is more than the number of independent variables, wwe get partial differential equation of higher order. Problem 1.1.1. Find the partial differential equation of : = az +by +ab by eliminating the arbitrary constants a and b. Solution: Given equation is sear thytab aGHAPTER 1. PARTIAL DIFFERENTIAL EQUATIONS ‘Differentiating (1) partially [Link] 2 Fnaspeo ) Differentiating (1) partially [Link] y 2 ye gad (3) ay Eliminating a and b using (1),(2) and (3), we get == prtqy+py, which is the required PDE. Problem 1.1.2, Find the partial diferential equation of = = 02+ tty +ab by eliminating the arbitrary constants « and b ‘Solution: Given equation is =a'z + Hy +ab a) Differentiating (1) partially ws:to aop=0 @ Differentiating (1) partially [Link] y a é eet @) Eliminating a and 6 using (1), (2) and (3), we get 2= pe tay + Pi, whichis the required PDE Problem 11.5. Find the poral diferente equation of : = ax + by + VaTFE diminating the arbitrary constants a and b. % Solution: Given equation is : snerttVeFR w Differentiating (1) partially [Link] 7 & eee ae ia Diflerentiting (1) partially w.:t0'y a: Papo a Eliminating a and b using (1), (2) and (3), we get ==patay+ VP+#, which is the required PDE. | Problem 1.1.4. Find the partial differential epuation of 1.1. FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 5 (28 + ab)(y2 +) by elimi- nating the arbitrary constants a and b ‘ Solution: Given equation Is = (2 + at\(y? +07) ) Ditlerentiting (1) pnetally we to 9 yh 4 tyne > p= 2ely? +b) 2) Differentiating (1) partially w.r-to y a: = (24+ b)2y = 9 = 2y(a? +0) @ oy Bliminating a and 6 using (1), (2) and (8), we get (2) x (3) = pa = dzu(2" + a)(y? +0) = pq = Axyz, which is the required PDE. Problem 1.1.5. Find the partial differential equation of (x ~ a)? + (y— 6)? +27 =1 by eliminating the artitrary constants a and b. Solution: Given equation is (=a (y-or rear ® Differentiating (1) partially w.r-to 2 2a ~0) +229 = 0— (e~0) = 2p ® Differentiating (1) partially w-rto y Ay — 0) + 222% = 0 => (yb) = 29 ) ay Eliminating a and b using (1), (2) and (3), we get (1) = (2p)? + (-29)? + 2 = 1 9 4" +g? +1) = 1, which is the required PDE. Pootla Problem 1.1.6. Find the partial differential equation of (x —a)* + (y—b)? by eliminating the arbitrary constants a and b. Solution: Given equation is (2 =a)? + (yb)? = sPeot? a a Differentiating (1) pastially w-to 2 Ax-a) cot? a2 = (2-2) = zpeotta °F 6 CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIONS, Differentiating (1) partaly w.r10 y 2y —0) = cot ade —> (v0) = 20008" @ ‘liminating o and b using (1),(2) and (3), we get (1) = ( pe + ay=>2, which ia the required PDE (Option 118. Find the portal diferetial equation of = = ae! sinby by eliminating Solution: Given equation is ae sinby Diferentating (1) partially w to 2 a : Be = abe sinby = p = abe sin by 2 Diflerentiting (1) partially [Link] y 0: = abe cosy => 4 = abe! cosy ° @ 1.1. FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 7 Eliminating @ and b using (1), (2) and (3), we get From (2) we get,p 2 Patan (2) = p= atan (%), which nthe required PDE, Problem 1.1.9. Find the partial diferential equation of z = a(z+y)+H(2—y)+abt-+e, ‘where x,y and t are independent variables, by eliminating the arbitrary constants a,b and Solution: Given equation is 2=a(z+y)+H(z—y)+abt+e a) Differentiating (1) partially [Link] 2 o morb—paatd @) Differentiating (1) partially [Link] y 0 (4) Eliminating a,b and c using (1), (2) (3) and (4), "we get Pa @ = sab = P-@ = 4%, whieh isthe roquized PDE. Problem 1.1.10. Find the partial differential uation of 2: = 2+ ¥. by eliminating the arbitrary constants a and b. Solution: Given equation is _2,e d= ate a) Differentiating (1) partially [Link] = Se = Forme (2)re I te io oS ee hee = Su 1= 3 6 ) Eliminating a and 6 using (1), (2) and (3), we get oe aes & and : oser-2 ree Ve W=2= +7 © => 2: = pz + qy, which isthe required PDE. @ Problem 11.11. Find the partial diferential equation of (1+ a2)= = (= + 01 +0) by eliminating Use arbitrary constants a and b Solution: Given equation is) t A(1 +.0?)z = (z+ ay +6) a) Differentiating (1) partially [Link] x é Mit at) BE = 22+ oy +6) = 11+ a?)p=(e+ ay +8) 2) Diflerentiating (1) partially [Link] y M4 NEE = 2s tay + Bla —> 201+ aig = ofc + ay +b) @) Eliminating o and b using (1), (2) and (3), we get (e, @~ 47a O—raenaifies)s eels] => 2=(p7 +9’), which is the required PDE. 1 CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIONS. | Le 1.1. FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS (Problem 1.1.12. Find the partial differential mation of 244 42 ferential equation of 24 84% = 1 by eminating the arbitrary constants a,b and c. Solution: Given equation is Differentiating (1) partially [Link] Ben+ Son Differentiating (1) partially [Link] y Ben + eng 0 Differentiating (2) partially [Link] y Pz az yas Pas] ~° s+ m=0 which is a PDE of (1), Remark:1 Differentiating (2) partially [Link]-z, we get Js APs + pt #* al'ae a sete] Snr ta ap=0 3(6) - (2) = which is also a PDE of (1) Remark:2 Differentiating (3) partially wrto y, we get 1 1p, a2)_g tit b+ 5 [35 +0] -0 +5 | 1) -@) = Serre] -5i=0 Syst +f - 9 =0 which is also a PDE of (1). Thus for the equation (1), we have three PDEs a @ @ “ © @ ) (8)i 10 CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIONs | Problem 1.1.18. Form the partial diferent uation ofall planes which are at con, stant distance k from the origin. | Solution: Equation of all planes which are at a constant distance & from the Origin takey the form iz+mytne=k a] where Pamgnt=l @ Difierentiating (1) partially w-rto = 14nd 091 4+np=0 @ Differentiating (1) partially [Link] y m+n2 2 0—2m+ng=0 @ Oy Eliminating 2, m,n using (1),(2),(3) and (4). () + (Cnp)z + (-ng)y tne = VEEP (pe ay+ 2) = kV Gat mae n(—p2 — ay +2) = nye aT z=prt+qytk/p+¢+1, which is the required PDE. 1.1.2 By eliminating the arbitrary functions Consider the equation s=u(2) @ wher fan wba fiction [1e) mems(2)- Dilceniating 1) paiaty ws = z-4'(2)( @) 1.1. FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS u Differentiating (1) partially [Link] y EO meteG) = pr +qy, which is the required PDE corresponding to the given function. Remark: This is o PDE of order 1. In general, if the given relationship connecting ,¥,2 has only one arbitrary function the resulting PDE is of first order and otherwise, ‘we may need to go for higher order PDE. Problem 1.1.14. Find the partial diferential equation of : = 2-+y-+f (zy) by eliminating the arbitrary function f. Solution: The given function ie z=rtyt flay) a) Diflrentiating (1) partially [Link] 2 SE=14+ ep p=140f a) @ Differentiating (1) partially [Link] y a Fai ss pa= t+ sev @) @) > p-1= sen) () 0-1 = 27a) Po Yas -e= (02 dy 1": pe qy~ =~ which athe required PDE, Problem 1.1.16, Find the portal diferent enuation of: = 1 (2) by eliminating the arbitrary function f Solution: The given funtion a 12) 0@) 1.1, FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 13 Problem 14.18 Find the porta iret eration of» = w+ 2 (24g) to eliminating the arbitrary function f. Solution: The given funtion ts vay (2 + tony) Oy Differentiating (1) partially [Link] x ae-0r (E+ 0009) (- Differentiating (1) partially wto y, we got 5a 2 +7 (2+ 80) (GG) anaes 2p 3 + ey) @ ) = -3r (+ nr) @ 1 r(b+ be) = 2 =2(_= (ons 2y02(—AE) = pe + qy =2y, which isthe required PDE, Problom 1.1.19. Find the partial differential equation of z= f()+ ly) by eliminating (he arbitrary function f and 9. Solution: The given function is 2 = Je) +0) a Differentiating (1) partially [Link] 2 as BE pe) = pa sree 2) & = sa) = pose) @ Differentiating (2) partially w..to y, we get Pe =0=5 1 =0, which isthe required PDE. yar Problem 1.1.20. Pind the partial diferentil equation of = the arbitrary fanction f and 9. Solution: The given fanetion is S(y)+e*a(y) by eliminating z= f(y) +eoly) a) Differentiating (1) partially [Link] © SE a gly)e? => p=atunet @ Difereotiting (2) partially w.r.t0 2, we get 2 «gyi er =p, hich ithe ied POEa CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIONS Problem 1.1.21. Find the partial differential oration of = = {(=)9(9) by eliminating the arbitrary function f and 9 Solution: The given function is == f(=)olv) Q) Differentiating (1) partially [Link] 2 z = olv)F'(2) => p= ovis") Differentiating (1) partially [Link] y, we get & = seu) => = ev) Q 1) Differentiating (2) partially [Link] y, we get & =F(2)o'y) = = Sed) “ Now, (z)o(v)f'(z)o'(y) + pq = 25, which is the required PDE. Problem 1.1.22. Find the partial differential equation of z = f(x-+ iy) +9(2—iy) where t= VT by eliminating the arbitrary function J and g. Solution: ¥% v3 ‘Differentiating (1) partially [Link] = z= f(z+iy) +9(z—- iy) Qa $= sermtole—w @ Differentiating (1) partially [Link] y, we get FeV e+ iv) ide) @) Differentiating (2) partially w.:.to 2, we get a: aa EFM) HS 2-8) > ra setWtee-W) — W) Differentiating (3) partially [Link] y, we get ‘ Pe _ am Fy Bp TPE + Fe) = =P ew) to"e-W) Now, rant => r+t=0, which is the required PDE. J. FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 15 1.13 By eliminating the arbitrary function ¢ from (u,v) = 0 Let u and v be two given functions of z, y and z, Suppose rela tle pt y ippose there exists n relation between ou») =0 We want to eliminate ¢ and form a differential respect to z, we get (1) ‘equation. Differentiating (1) partially with ° @ @) 20) a9 #|[5]- ey) Lav. 2% 4 0 and 40, Therefore the above Here fumetion of u and v so diva id v so that z sytem of equations has a non-trivial solution & where uy = $2, ty = $8, vs = SE and uy = Remark: Note that when we compute the partial derivatives ts, tp, Us; tyy 2 should be trated as a dependent variable which is a function of the independent variables 7 and y. JProblem 1.1.23. Find the partial differential equation of oz +y+ 2,22 +92+=2)=0 by eliminating the arbitrary function ¢. Solution: Take u=2+y+zand v= 27+ y2+422 Stay, aay goite Baits bu ov Oi oe OY ay tae Bette Fat 29 ‘The PDE is given by yy Ltp_ztpe Thay ta => (y—2)p+(z-2)9=2-y16 CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIONS i in the form Pp + Qq = Remark: Note that the partial differential equation is R ‘where P.Q and R are in general functions of x, y and 3. Equation of such form is calle Lagrange’s first order PDE. 1.2 Solution of Partial Differential Equations Consider the example of forming PDE from the function artby+ab « ‘The corresponding PDE is, pet ava @ ‘That is (1) is a function which satisfies the PDE (2). Also oetytse ty te) @) is a function satisfying the PDE (y-2)p+(2~2)9= 2-9 @ ‘Such functions are called solutions of the corresponding PDEs. That is, a function of the form Gz, 4.2,0,8) =0 © (u,0) =0 © where u and v are functions of z, y, = satisfying a given PDE is called a solution of the PDE. Note that, a solution may or may not contain arbitrary constants or arbitrary func- tions. Tn (1), suppose we give specifi values a = 5,b = 7 then z = Sx + 7y +35 is solution of the PDE (2) and again note that = = —ry is also a solution of (2). That is, a PDE can have different type of solutions. So similar to the case of ODE, we need a proper classification of solutions of PDE for a systematic analysis of PDEs. ‘The solutions of PDE can be classified into the following categories, (a) Complete integral(complete solution) (b) General integral(general solution) (©) Particular integral(particular solution) (@) Singular solution 1.2, SOLUTION OF PARTIAL DIFFERENTIAL EQUATIONS 7 incl Solution of « PDE having maximum number of arbitrary constants which s equal to the number of independent variables is called the complete integral The solution (1) is n complete integral of (2). But the solutions » = mz + Ty 4 4a oF == dz +6y +24 are not complete integrals of (1) A solution having maximum number of arbitrary functions solution of the PDE ~ alled general The function == Sz) +9(y) @ is the general solution of ‘ # ® ‘Also note that rertty o is a solution of (7), but it is not a general solution of it, Note that the solution (9) is obtained by giving the specific values f(z) = x and g(y) = y in the general solution. Also the function Ae + by +4b (10) is a solution of (2) obtained by giving specific values to a = 4 in the complete integral (1). These solutions belongs to the class of particular integrals. That is, a solution of a PDE obtained by giving specific values to one or more of the arbitrary constants or arbitrary functions in the complete integral or general integral is called particular solution. The solutions ar+dy+4o, and ay =3r+4y+12 a2) are particular solutions of the PDE (2) ‘There is ancther type of slation of PDE which has few practical applications but has afew theoretical interpretations. "Those solutions ae calla singular solution. More caly rr tolution of a PDE having no arbitrary constants or arbitrary functions and which cannot be obtained from the complete integral or general solution ty alving specific values to the arbitrary constants of arbitrary functions is 2uea ngulas solution of the PDE. We shall lustrate procure to compute these solutions in the succeeding pages. 1.2.1 Solutions by direct integration z y lerivatives, then In a PDE if the dependent variable = occurs only in the form of partfal derivatives, t the PDE can be solved by direct integration. We shal illustrate this through the following examples.FERENTIAL EQUATIO} 18 CHAPTER 1. PARTTAL DIF! NS # 7 ay gy ZE-0 J Problem 1.2.1. Slee (i) Bao wy Bo on #3 Pz to) F=0 w am79 Solution: (0 Tempra POE oF =o, Siting wih rope fos mee == 70) whe Sly) is an arbitrary function of y. 0. imegating with respect to y, we get 2 = J(2), where (Tegra row f(z) is an arbitrary function of 2. (i) Teepe roe 2 <0 = (2 we get 32 = f(y), where f(y) iam arbitrary faction of y- Integrailag again with respect to 2, we get z = 2/(v) + ay), where f(y) and g(y) are arbitrary functions of y. @. (iv) The given PDE is 55 o a (2) = 0, Integrating with respect to z, y = 2 (#) = 00. Integrating with respect to a y, we get 22 = (2), where f(z) is an arbitrary function of z. Integrating again with respect to y,me get 2= yf(z) +9(z), where f(z) and g(2) are arbitrary functions of z. (0) Tre sre rob Ze wo > B(2) 0 teonting with raps to a we get: = f(y), where f(y) ia an arbitrary function of y. Integrating with respect to v, we get z= f(y)dy + o(z) = F(y) + 9(z) is the required solution, where F(y) and g(z) are arbitrary functions y and x respectively. i v Problem 1.2.2. Solve hy ae _sifar Iteating wt rope o zw get = SCE) 4) ee fy) na rity function of. Intertig wth resp oy, we et cooler + 2 TOL tayty ote) _ 1.2, SOLUTION OF PARTIAL DIFFERENTIAL EQUATIONS 19 is the required general solution of the given PDE, where f(y),and g(z) are arbitrary functions of y and x respectively om au /Problem 1.2.3. Solve > Solution: The given PDE is Oy ad = = 28)-- inz + f(t), where J(¢) is an arbitrary inet f s(t)dt+9(z) i the Iegating with pet, wget B= function off. Integrating with respect to f, we get u=—e required general solution, where f() and g(2) are arbitrary functions t and z respectively. a N\A _ aH) =sin(zy) SoHE. 4 2), where (2) isan arbitrary Pz roblem 1.2.4. Soe PF = sin(zy ” en sae Solution: The given PDE is a Fave) Integrating with rxpect toy, we get $= function of. Integrating gaia with expect toy, we gt 2 = — 82 + yf) + of), isthe reed general ection ofthe given PDE, where f(z) and fc) are arbitrary functions of z alone. 2siny, when x = 0 and ez oz roblem 1.2.5. Solve rainy for which 2 Problem 1.2. sinzsiny fe a 220 when y is an odd multiple of 5 Solution: The given PDE is Ze anasioy > Z(%) -aosany Integrating with respect to z, we get 5 = —coszsiny+ fly), where f(y) is an arbitrary function of y. S be Given that 2 = —2einy, when 2 = 0 and so ay ’ —2siny = —siny + f(y) => f(v) =—siny 2 «cen zsiny —siny = —siny(1 +0082) “OyIE involving two inde satisfying the given PDE will be its ats a or bor both, we get the particular 1.2. SOLUTION OF PARTIAL DIFFERENTIAL EQUATIONS a Then (1) becomes Gz, y, 2,4, 6(a)) = 0 3) Differentiating (3) partially with respect to a, we get ag 0 “ Now we eliminate « between (3) and (4). The resulting relation gives the general solution. Geometrically singular solution is the envelope of two parameter family of surfaces G(z,1, 2,4,0) = 0, the complete integral. So to find the singular solution of the given PDE we can use the following procedure. Differentiating (1) partially with respect to a and b separately, we get ag a ° © © Lo 2% Eliminating a and 6 using (1),(5) and (6), we get the singular solution of the PDE. ‘We shall illustrate the procedure using some simple examples. We found that the PDE corresponding to the function ematr+ Py tab a) r=petqvt VR @) So (2) isi the complete integral of the PDE. To find the general solution, we take = Ha) ° so that 2s atz+ (ay + ad(a) @) Thea Fg + raz +24(a)6'(a)y + o(0) +.00'(a) =0 0 a Eliminating a between (4) and (5), we get the general solution ofthe PDE. To find the singular solution we form the equations 5 = 0 and 5; = 0. ac es = 6) Fe =0 = tor +b=0 @) no = ty +0-0 m Now we eliminate a and 6 using (1), (6) and (7). (©) = b= -2ar (7) = 2y(-20z)+a=0 => ~Ary+1=0 ‘Thus 4zy = 1 is the singular solution of the PDE.i PDE 1.2.3 General form of First order yaiables = and y can by ‘The general form of fret order PDE having two Independent expressed as Pleo BE + Qesnigg + Rese = 59) where P(e 9),Q(69) lov) ste) ae functions of z and vo: cones OF 5 sbeent from the frst order linear PDE, we can ened aa Or oe el Vere Sty tretial equation with i sole the PDE using methods of solving first understanding that the arbitrary constant is For example consider the PDE function of one independent variable. mptten? (ay Here ¢ = 2 is absent and the equation (1) is comparable with the ordinary linear ferential equation £ + Play = Q(z) and Q(z) =. Hence the integrating factor is Bieee es = Hence the solution is given by is the required solution ofthe PDE, where f(y) is «function of y Consider another example. igte=0 ‘Tofind the solution we can make se ofthe method of separation of variables. a, Integrating the above equation with respect toy, we get , logz-+logy = f(e) <> log(y2) = (2) land $0 the solution ia y= ol) (2) IAL EQUATI HAPTER 1, PARTIAL DIFFERENT on 1.3, LAGRANGE'S LINEAR PARTIAL DIFFERENTIAL EQUATION 2 1.2.4 Linear and non-linear PDE (Classifying partial differential equations into linear and non-linear PDE ls a most signifi- cant step towards finding solution methods for a large class of partial differential equations, Similar to the definition of linear ordinary differential equation, we define it for PDE. A partial differential equation in which the power of dependent variable parial derivatives are either 0 or 1 and there is no term containing product of dependent variable and its partial derivatives is called a linear partial differential equation. Otherwise it is called non-linear partial differential equation. The partial differential equations akira atte ou, Bu Batt ge 7 0V) ae z7 are examples of linear partial differential equations where as the following partial differ- ‘ential equations Pu Oo az)? ® +2 from Method of reping: Tn he math of oping we or a qution by oping Sither F080 ad Sox at © sch tat ee variable auctor can be removed om both, Then ig som ntod slut of rdinay deren ustows we ca the felon de ‘Method of multipliers : In the method Properties of ratio and proportion 1 ‘7h then for any choice of numbers tm, we have of multipliers we make use the following = fax +may+ nay BF mb, + nb, 1.3. LAGRANGE'S LINEAR PARTIAL DIFFERENTIAL EQUATION 25 Care should be taken to choose suitable l,m,n which are constants or functions of = such that the denominator [P +mQ +n vanishes or is proportional to either P. a ‘a combination while applying this to & = i oF Qt Ror a combination while applying this to $= = = S, We shal istrate the above procedure while solving the following partial differential equations. Problem 1.3.2. Solve xp + yq = :? Solution: The given PDE is 2"p-+ y?q = 2°, Auxiliary equations are dr _dy_dz se 3 Ting tet ro tn, SF = a st in inerig me 2 Solution: Given equation is “=pt2sq—y! => yap + x2x a, ao we vst equations ae = y estos ae HE é “a ng th fit i rt, A= SB my Pra e Integrating, we get Hence the general solution is 4 P= aes) Oo26 CHAPTER |. Problem 1.3.4. Solve p—¢= lost +¥) qe logle+y) | = first order PDE is p~ Solution: The given e's ner Ausilinry equations are Taking the Bat two ratios Integrating, we get z= -y+a => 2+¥ i a T 7 foga — hades = as Taking the first and last ration, = Sere y) Inegrating, we get ogar = 246 => zlosle +9) Hence the general soition et neiate tis) 0 0 et" Problem 1.3.5, Solve pz —qz = 34+ (z+v)" Solution: The given Lagrange’ fist odes PDE ie 72 — 02 a Auxiliary equations are F = Y= 5G Tang te tw rt = Integrating, we got r= -y +o =o z+ y=0 = de=-ty de de T we first and last ratios, = = —— “= . aking the Bist and last ration, = a ds 25 Fea Integrating, we get = 5 log(=? +04) +o <> lols? +(z + y))] -22 logy =loga: —> Ym e de Taking the multipliers = .2,m = yn = x, we get IP + mQ + nh = xf +12 — zy - 227 20 > rdet ydy + edz =0 Integrating, we got op op zigty-o Peyperen Hence the general solution Is o(2rr+2) pt yet —22)) = (2) Payet)aoo Problem 1.3.7. Solve x(y’ Solution: Given equation in 2(y! - <4)p + yz? — 2*)q Ausiliary equations are ‘Taking the multipliers! = z,1m = y.n-= 2, we get Pi mQ tak = 2H ~2)+ 72-2) +2%2"— 7) =0 = ride + ydyt edz =0 Integrating, we get 11 ‘Taking the multipliers mm tin = 2, wo get IP +mQ+nk = (y ~ 29) + (2-21) + (P-7)=0 a Lact bys lueno Integrating, we get log.r + logy + log 2 = log => log zy: = log ayz=b Hence the general solution is o(erv sean) =0 9 Bre ae(a)CHAPTER 1. PARTIAL DIFFERENTIAL EQUAT Yo, 28 2-r)q= 2-9) Problom 1.3.8, Solve 2%{y —2)p-+y"(E- 714 ae Sobution: Give eqnatian a 2¥%p—a)p 422 = HUE H Ausilinry equations arc ot Ayo *¥e-) Fen) J edna, weset ‘Taking the multipliers = 2) +ye-2) 428-9) IP+mQ+nh=29 Integrating, we get log.r + logy + log: = loga => [Link]= = toga Taking the multipliers (= IP-+mQ+nR=(y-2)+(2-2)+ (2-9) =9 = feet bays Seno litegati, woh vay 112) oon bets!) a0 o anea(teted step ey EE we era? wy +(y—z)p + wz - 2) Problem 1.3: -») Solution: Given equation is Ausiliary equations are ds ty de 0-9) We-2) e—D wwe get Ting he malities yn IP + mQ+nR= (y~2)+(s—z)+(e-y) =0 Edy ed = fdr tidy + tae pt ywte Integrating, we get loge + logy +g: =loga <> lossy: = loga nr=e 1.3. LAGRANGE'S LINEAR PARTIAL DIFFERENTIAL EQUATION ‘Taking the mmultiplicrs = 1,m = Lyn = 1, wo got IP + mQ +n = aly ~ 2) + (22) + 2(2-y) =0 = lds + Ldy + 1d: =0 Integrating, wo got rtytesh Henee the general solution is Sori e4as) 0 o myeno(cree3) = 2 )\p + 2ryq = (2? = = 2)p + 2zyq = Bre, Problem 1.3.10. Solve (2? ~ Solution: Given equation Is Auailiary equations are Taking the lust (0 ratios v Integrating, we got logy = log: + loga => logy = logaz Also det ydy + 2de 8 = xy? — 228 + Day? + 2x2? = Ede tydy + sie a+ ry? + 23! zdutydyt zde 7e+y +2) dy _xde+ydy+zdz Dey ~ ae + y+ 2) dy _ xdx + 2ydy +2 Bree dy dette +2) yo eye Integrating, we get logy = log(s? + y? + 24) + Loney faye | y13. LAGRANGE'S LINEAR PARTIAL DIFFERENTIAL EQUATION a1 Integrating, we get Foy F+h+ gee Peyttab ‘Hence the general solution is o(f-ae- 2.24742) =0 on ump-2=0(247+2) Note: In the above problem if we take u = y2, then au um 8 dos ay + Fede = ody yds Problem 1.9.12. Solve (y +2)p+ (s+ 2)=2+9 ‘Solution: Given equation is (y+=)p+(=+2)9==+¥- ‘Auxiliary equations are a dy de Then Integrating, we get log(+ - v) = lost - 2) + loge z-v_, ye Also dey _ de dy tae voz Wetyt2) de— dy _dr+dy+ dz Foy tte Integrating, we get =2log(s — y) = loge +y-+ =) +06 (e+y42le-WP=h +22) +2(zy— 22) =032 CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIoy, Henco the general solution i o(2EBtervs tes?) =0 oS (e+e -v») yand v=z+y+z, then ‘Note: In the above problem if we take =~ Problem 1.3.13. Solve (2? ~ yz)p+(y? - 2#)0= 27-70 Solution: Given equation is (x? - yz)p+(v - =)q = 7 ~ 7. Anailiary equations are Then aos da @-w)- 023) P=) - Pa) a Gere” OFA FW) de—dy tyne (@=We+u+2) W-Die+0+2) me Body _ dy de yo yas Integrating, we get og(z ~ y) = logy — 2) + loge, ay me Also re aetdytd: de tydy tide PEP+ P= yp Beets dee setdvtds zdz+ydy-+ ede BFP Fa ay- Bg” Wyte Se = de dy + dy = MEH VaYt ede tyse > (ety sNde+ dy ds) = 2ds-+ yay zdsloteprating, we get G@tvee at ya Ft tte SS ytetynd Hence the geueral solution is 4 sesetn) <0 rain o(atete) 14. EXERCISE i Problem 1.3.14. Solve (2+ y)zp-+(z—y)zq= 2 + Solution: Given equation is (+ yp+(— vga 2 Awailiary equations are este viae +" a Gta: Gye Bae Taking the Brst two ratios aedx ~ yd ~ rdy ~ydy 2de—d(zy) — ydy =0 Integrating, we get F-a- ba P-2y-~=a Taking the multipliers IP+mQ +n = r(2-+y)s~ylz~ yz —2(a? +4) =0 = rde~ydy—2d2=0 we get Integrating, we get Hence the general solution is o(2-29-¥, 8-7 — 1.4 Exercise 1. Find the partial differential of the following equations by eliminating he arbitrary constants a and b. () = (+ ayy to) (i) zsartay th (ii) 2= (240+ (ys 62 Solution: 42 =p +q +42CHAPTER 1. PARTIAL DIFFERENTIAL EQUATIon, Solution: q = pr+7? Z+¥4ie1 (a) Find the differentia! equation ofthe family of the spheres tres lies in the 2y~ plone. : va Solutions. Hint See the problem 418 (¢-0)*+(y- BP +2? = gy 20+ e+i=a (vii) Find the aiferential equation of all the spheres whose centres lies on = axgg, Solution: Hint : See the ilustrative example. (Gx) Find the partial dilerentil equation of allright circular cones whose ayy coincide with 2 axis. 2. Find the pactn! diferential of the following equations by eliminating the arbitrary functions of radius "a whoa w ==s(2) Solution: px+qy=0 (i) z= J(sinz + cosy) Solution: psiny + gcosr=0 (ii) = f=) Solution: py-+ gz =0 (iv) z= f(z+ay) + 9(2~ay) Solution: r= a () 2= 27 (2) +002) (4) = f(20+y) + 9fdz—y) Solution: r+ 1 = 6! ere(e— 9) Solution: p+q = nz (wit) 2 = (2+ ole) 4. Find the partial dleretal of the follwing uations by lining the arb function . . 7 “a (wi) W) fl +02+24,2 27) =0 (i) (a? +7 + 2.292) =0 ) o(f9-#) =0 Ps 4 Sone 22 sine andy Solution: == -yeovr+a(s)+ [ fuyey eee 15. NONLINEAR EQUATIONS OF Hiner onDER % soars BH 2 satan) 6 site Fans Solaed? = nna taf) ct oe 1 Sate (PE —eteny or wih «20 when z= 0and 20 whey 0 Solution: z = (1—e-*)siny 2 oe gay 2 sine Baar ey: Bae . Solution: = = 327+ 32y— 2)? +e 9. Sve (ma — my) SE + (na — 1) 10. Solve (y+ z)p~ (2+2)9=2-y LL. Solve peot.z + geoty = cot = 12, Solve py + 9/9 = V2 13, Solve p-+3q = Sz + tan(y —3z) 4 p 14, Solve 24 ny 15, Solve (x? — y? ~ yz)p + (2? - y? — =x)a (r-y) s 1.5 Non-Linear Equations of First Order For solving non-linear equations of first orcer we use Charpit's method. This is a general ‘method for finding complete integral 1.5.1 Charpit’s Method ‘A first order partial differential equation with two independent varinbles x and y and dependent variable z can be expressed as F(z,y,2,p,q) = 0. Since z is a function of z and y, we have a Fatt Fad = ple tay Suppose we are able to find another relation $(z,y,2:p.4) = 0, then we will be able to solve the above equations to get the value of p and g. Putting these values in dz = pd + qdy and integrating we get the required solution provided itis integrable.= 1. PARTIAL DIFFERENTIAL EQUATIONg (ay (13) a3, ay : Uiing the abore four eqeatizs, we ciminate 2 berween the frst to eqwations ang (cSemcmte & between the last equstions (of the four equations). we eet CES) EE SES S23 0 Noting that = Ee = 2 and adding above two equations, we get (on further rearranges) OF 0 OF 0 (OF OF) be (OF Oo Oa "a Tm) a \ az PO: (OF ,@F) 00 (OF . 8F) do _4 op 7G: ) dq ~ an (Coxsparing this with Lagrange's cer equation, we can se that this is Lagrange’: linear equation with =, y,= 9,9 25 tndependext variables cod 6 as a function of these variables Jes axiary equations (or subsidiary equations) are given by e d= pir +ody od integrating we pet the solution of the given difirental equations. We of the simplest of the aqeations from the safely eomannee i Problem 15.1 Using Qharpt's metho, find the complete integral of q= 247 ‘Solution: lt Fe927a)=3-4=0 im) 15. NON-LINEAR EQUATIONS OF FIRST ORDER Charpit's subsidiary equations are aly ds BF okt ar a =p) From the fourth fraction, we get dp=0 = p Substituting this value of p in (1), we get q-= 3a, Putting the values of p and q in d= = pdr +.qdy, we get det ady dz + 3a dy Integrating, : = ax + 3a"y +b, which is the complete integral, @ and b are arbitrary constants. Problem 1.5.2. Use Charpit's methods lo solve q+ px = Solution: a Le Fla,y.2.p.¢)=9+2- 7 =0 ‘Chaspit’s subsidiary equations are de dy 4p 37 oF here F with suffixes denote the corresponding partial derivatives of F with respect to the respective euffix i _. 2 p+ HO) 040) From the last expression, we get dg =0 => g=0 Substituting this value of q in (1), e get atpend = Putting the values of p and q in dz = pdz+ qdy, we get