nw ; Ips ¢ Regenerating Mathematics An ISO 9001 : 2008 Certified institute Sudea's Proneer Insiidate off a a ei a M.SC. Ent. (DU) A High Quality Study Material or Higher Level Exam for V.G. f P.G. Students Partial Differential Equation A USES CCC M OO ME DEEL OL EL 7Chapter 1: Classification and Formation of PDE. Ld 12 13 14 15 Partial Differential Equations (PDE), Order of a PDE .. Degree of a PDE jassifcation of First Order PDE 14.1 Linear Equation 1.4.2 Semi-Linear Equation ... 143 Quasi Linear Equation 1.44 Non-Linear Equation Formation of PDE 1.5.1 By Elimination of Arbitrary Constants 1.5.2 By Elimination of Arbitrary Function Chapter 2: Linear PDE of Order One 2 22 23 24 25 Chapter 34 32 33 34 ‘Lagrange’s Method .... 24.1 Working Rule For Solving Pp+Qq= R By Lagrange’s Method. 2.1.2 Four Types of Problems Based on Pp+0q=R .. Cauchy Problem for First Order Quasi-Linear PDE, Existence and Uniqueness of Integral Surface Passing Through a Given Curve wna. 12 3 Surfaces Orthogonal ta a Given System of Surfaces .. n Independent Variables ‘The Linear PDE wit (on-Linear PDE ‘Types of Solutions... 3.141 Complete Integral (C-1). or Complete Solution (C.$.) a 3.1.2 General Integral (G.L) or General Solution (G.S,) ..... 3.13 Singular Integral (S. 1) or Singular Solution (S.S:) Compatible System of Fits Order Equations Charpit’s Method .. 34.1 Working Rule While Using Charpit’s Method ... 3.42 Special Methods of Solutions Applicable To Certain Standard Forms35 36 Chapter 4: 4a 4d 43 44 Chapter 5: Curves & Surfaces in Three Dimensions ... ~ Calich’s Method of Ciariicteristics Tor Solving Non-Litea P.D-E- Linear PDE With’ Constant Coefficients and Equations With’ Variable Coefficients ‘Homogencous and Non-Homogeneous Linear Equations witi Constant Coefficients ‘Working Rule For Finding Complementary Function ... = ‘Method of Finding Particular Integratinig =... PDE with Variable Coefficients Reducible to Equations with Constant Coefficients... 44.1 Method of Reducing Euler-Cauchy Type Equation to a Linear PDE with Constzat Coeff cheMts nnn 44.2 Working Rule for Solving Euler-Cauetty Type PDE... Classification of Second Order Partial Differential Equations, Characteristic Curves and Reduction to Canonical FOP monn Chapter 6: Heat, Wave and Laplace Equation 61 62 63 64 65 66 Introduction Method of Separation of Variables unm Some Important Equations un Solution of the Heat Equation Solution of the Wave Equation ... Solution of Laplace’s Equation ann om 646.1 General Solution of Two-Dimensional Laplace's Equation 66.2 Dirichlet Problem in A Rectangle... Chapter 7: Green’s Functions 1 Construction of Green’s Function Assignment Sheet-1 Assignment Sheet-2 ..« Assignment Sheet-3CHAPTER 1 CLASSIFICATION AND FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 1.2, Partial Differential Equations (P.D.E.) ‘An equation containing one or more partial derivative of an unknown function of two or more independent variables is known as a partial differential equations. we list the following examples x differential equations: : SS 1.2. Order of A Partial Differential Equation ‘The order of a partial differential equation is defined as order of the highest partial derivative occurring in the partial differential equation, In above examples of partial differential equations (1),(3), and (4) are of the first order, (5) and (6) is of the second order and (2)is of the third order. 1.3. Degree of A Partial Differential Equation The degree ofa partial differential equation is the degree of the highest oder derivative which Gccurs in it after the equation has been rationalized, ic. made free from radicals and factions so far as derivative are concerned, In above examples of partial differential equations (1),(2),(3), and (4) , (6) ate of frst degree while (5) is of second degree. Notations: When we consider the case of two independent variable we usually assume them to bex, y and assume z to be the dependent variable. We adopt the following notations throughout the study of partial differential equationsAde “-Classification of First Order Partial Differential Equations 1.4.1, Linebr Equation ‘A first order partial differential equation is said to be a linear equation if itis linear in p.g and 2 , ic, ifitis ofthe form P(x) p+O(x»)q=R(x,y)2+5(x,9) Examples: yp—xq=xz2+x and p+q=z+3y ‘Semi-Linear Equation A first order partial differential equation is said to be a semi-linear equation if itis linear inp and’g and the coefficient of P and,Q the functions of x and y only ie, ifit is of the form \ P(xy)P+O(%y)q=R(ay.2) S wa WY Examples: e* p-yxq= x2" and yp + xq =~— — 2 LW \ \ Quasi Linear Equation ~~ ‘A first order partial differential equation is gaid Y.be S\Quasi-linear equation if itis linear in p and q , ic. ifit is of the ~ P(x,y,2)P+O(%»2)9=R(% yz) KT Examples:(?+)p—9q=2 + ere Non-Linear Equation ~ Partial differential equations of the form f(x,y,2.p.9)=0 which do not ‘come under the above three types are said to be non-linear equation. Example: pq=2, p2-+g?21 and ‘2p?+y%q?'=2? are all, non-linear partial differential equations: Formation of Partial Differential Equation By Elimination of Arbitrary Constants’ Consider an equation F(;y,2.08)=0 Al) Where a and & denote arbitrary constants. Let 2 be regarded as function of two independent vatiables 1x and y’. Differentiating (1) w. r. t xand y partially, we get OF | oF Fp ang FE Q a Oe a +5 (2) Eliminating two constants @ and 6 from three equations of (1) and (2), we shall obtain an equation, of the form f(x,y,2,p,q)=0 which is, partial differential equation of the first onder. Ina similar manner it can be shown that if there are more arbitrary constants than the number of ‘independent “variables, «the above -procedure’of elimination will give tise to partial differential equations of higher order than the firs. eae“Example: Form the partial differential equation by eliminating arbitrary the * constants a&b from z=ax+by-+ab, Solution: Given z=ax+by+ab (1) partially with respect to x and y we obtain Substituting p and q for a and b in equation(t), we get the required partial differential equation as z= px+qy+ pq SN ‘Three Situations May Arise Situation I: When the number of arbitrary const Sea thin the number of independent variables, then the elimination of ants usually fives rise to more than one partial difrenial eating one. For Example: Consider 7=ar+y ..(l) me Where a is the only arbitrary constant and “xy Yb two independent variables. OAL Differentiating (1) partially w. r.t. ‘x’, Weget 2) a t= Differentiating (1) partially w. r. oe < o ey Eliminating a between (1) vai(2)+7 Since (3) does not’ contain arbitrary constant, so (3) is also partial differential under consideration. Thus, we get two partial differential equations (3) and(4). / Situation I: When the number of arbitrary constants is equal to the number of independent variables, then the elimination of arbitrary constants shall sive rise to a unique partial differential equation of order one. For Example: Eliminate aand 6 from az+b=a"x+y Differentiating (1) partially w.r. x and y , we get ym) Fliminating a from (2) and 3), we have 5 1, which is the unique partial differential equation of order one. Situation III: When the number of arbitrary constants is greater than the number of independent variables, then the elimination of arbitrary constants leads to a partial differential equation of order usually greater than one.thy texy Differentiating (I) partially w.r.t xand y ,we have fase ox & Saber ay Differentiating (2) partially w. rt. xand(@)w.r.t. y, wehave ) eh Differentiating (2) wat y, we have Now, (2) and (3) , 33{Z)-ecvey and f 2 {Bf 8 porate ae } (By using (1) and (6)) Thus, we get three partial ‘differential equation given by (4), (5) and (7) which are all of order two. . By Elimination of Arbitrary Function 4 Let 4(u,»)=0, where w and v are functions of x,y and 2. We treat 2 as dependent variable and xand y as independent variables so & =0 and =| oy Differentiating (1) partially wart. x, we get & , & a), a6( a Gye ee) Hx (Be p28) 2h Bula" Gz) Ov\de “Me ee Sege ) (% y tala x ou dv _ au! Be ov Geer Ox RN a Oy Oy Or ‘Thus we obtain a linear partial differential equation of first order and of first degree, ie. in pand q : Example: Form a partial differential equation by eliminating the arbitrary function ¢ from 4(x+ y+z, x? +y*—z*)=0what is the order of this partial differential equation? Solution: Given (x+y +2,32 +y°~2)H0 ) Let u=xhy-42 and vax +yh“2? --Q) Then (i) becomes: (1,v)=0 (3) Differentiating (3) wer. ¢ x partially, we get ald) from (2), from (4) and (5), ()or)-2(%4)o-m)=05 (# ; nim 7 a) (3) Again, differentiating (3)war.t, y partially, we get v8) Met) ul ay a ey %% Svg)sa(S)o-x)-o.ty 9 (3) tym) 06) (1+4) Eliminating ¢ from (6) and(7), we obtain (x= P2)_(-w) aS (ter) Gea) or (I+4)(x~ pe) =(1+ P)(9~27 Dar (r+dacbe xy ‘whieh i the desired partial differential ee st order. Example: Form the PDE by cli He wbitrary function from. z= f(x+it}+g(x-it), where i= FAN Solution: SAY Given 2= f(x+it)+e(x-) — ll) Differentiating equation (3) twice partially with respect to x and’, we get Zap (evi)e'(ei) & =f"(utit)+e'(x-it), , 2) if BS S" indicates derivative of f with Tespect to (x-+it) and g' indicates erivative of with respect to(x-it). Also, we have Saif (esit)- e'(x-i)] 22 Sea (p (asin eet) ‘From Equation (2) and(3), we at once, find that tz Oz wea” which is the required PDE. Note: We observed that by eliminating arbitrary functions, we always | produce quisilinear partial differential equations, only. However, we get both quasi-tinear as well as non-linear partial differential equations when we |} eliminate arbitrary constants.CHAPTER? LINEAR PARTIAL DIFFERENTIAL EQUATION OF ORDER ONE : Lagrange's Method Theorem: The general solution of the quasi-linear partial differential equation Pp+04=R is Hus)=9, ee ‘ sme eh inependent soins of 5 = = ~~ Where ¢, &c, are arbitrary constants and are 2% ~ aS 2d. La the standard form wo Pp+04=R wll) Step 2: Write down Lagrange’s auxiliary equations for (1) namely, (2) Step 3: Solve (2) by using the well known methods. Let u(x,y,z)=4, and (2,942) =e. bet liferly independent solutions of equation (2). Step 4: The general solution (or integral) of (1) can be written in one of the, following three equivalent forms (19) =0 oru=9(s) of =6(u Four Types of Problems Based on Pp+Qq=R ee (i) Suppose that one of the variables is ether absent or cancels out Geom any two fractions of given equation (1). Then an integral can be obtained by usual methods. The same method can be repeated with another set of two" fractions of given equation (1). Working Rule For Solving Pp+Qg=R nge’s Method a F Step 1: Put the given linear partial dferehtighequation of the first order in | F$an, ire) Sra Hour Kaa Near ELT, Ney Del i006 Pks (LI) HAUTE, Cat OUR R SINT AEA Ea ifsdoncaden zag: Wea: wredoacsdcecomSolution: Give 2 2 poms y ‘The Lagrange’s auxiliary equations for (1) are dy Gaye? ‘Taking the first two fractions of (2), we have Para ytdy or ae-y*dy=0, din yiak or xde—zdz=0 Integrating (5), (1/2)x? ~(0/2}22 =e From (4) and (6), the required general integral 4(2 -y.x*-2")=0, @ being an arbitrary Y ‘Type & LA POR QRY ~ Suppose that one of integral of equation (1) known by using Rule-L and also suppose that another integral cannot be obtained by Using Rule-I, Then ‘one integral known to us is used to find another integral as shown in following solved example. Note that in the second integral, the constant of integration of first integral should be removed later. Example: Solve p+34=5z-+tan(y~32) Solution: Given p+3q=5z+tan(y—3x)" (1) ‘The Lagrange’s subsidiary equations of (1) are & & 13) Se+tan(y—3x) ‘Taking the first two fractions, dy—3ek=0 Integrating (3), y-3x=q ae Using (4), fom 2) we get = 5 Integrating (5), x-(1/5)log(Sz+tanc,) =(1/3)ep or $x—log[ Se +tan(y~3x)] =p, using (4) from (4) and (6) the required general integral is ‘Sx—log[ 52+ tan(y~3x)]=6(y-3x), where 6 is an arbitrary function. (FC er Maas aN LT New Da ONG PUTRI, Ca SIEGAL FHLGTIN RESET nal lafioicaden can: Webi wnpeaeuca zo&, (0) POR 0) Let R,Q and R be functions of x,y and z. Then by a well known principle of algebra, each fraction in (1) will be equal to Pads + Oydy + Ryde Q RPSQO+RR IFRP+Q0+RR=0, then we know that the numerator of (2) is also zero. This gives Rér+Qdy+Rdz=0 which can be integrated to give t4(%y,2)=4. This method may be repeated to getanother integral (x,y,z) =¢2-R,QoR; are called multipliers. As a speci ie can bie “by use bbe constants also, Sometimes only one integral, i multipliers. In such cases second integral shauld ‘using Rule-| 1 or Rule-2. Yrs. Le}. Example: Solve {(b~<)/ a}. moh Solution: Given (o-oo) “The Lagrange’s ausiliary equations of (1) aS oS Coe Cae (a-by we Choosing x,y, a8 mult ie ach OS ing x,y, ip! eoree) adr tbydyteade —_ ade + bydytexde Bz[0-a+(c-a)+@-B) 0 acd +ydy +cat =0. or Dau Dbyy + cede =0 Integrating, ax? by? +2 =, .Q) gait, choosing ar,by,cz as multipliers, each fraction of (2) 2 Pade syd tected _ abrdetBydy +Pade [aloe Bo =a) + le — Integrating, atx? +5°y? 4022? =e, 4) Fron (3) and (4) the required general solution is given by (ax? +by? +e2?,a?x? 47y? +72?) =0, where ¢ is an arbitrary function, () Let R,Q, and R; be functions of x,y and z. Then by @ well known principle of algebra, each fraction of (1) will be equal to Aid + Ordy + Redz‘Supposé the flamerator of (2) is exact differential of the denominator of (2) ‘Then (2) can be‘combined with a suitable fraction in (I) to give an integral However, in'some problems, another set of multipliers P,Q and R are so cchosen that the fraction Badr + Ondy + Ryde 0 PP+O0+RR is such that ts numerator is exact differential of denominator. Fractions (2) and (3) are then combined to given an integral. This method may be repeated in some problems to get another integral. Sometimes only integral is possible by using this Rule-4. Jt such cases second integral should be obtained by using Rule-1 or Rule-2 of Rule-3. Example: Solve (y+2)p+(z+x)q=x+y Solution: Here the Lagrange’s auxiliary equations are, dc dy ae Ss yee atx xty Ve Choosing 1,-1.0 as multiplier, each fraction OF( . 4-9) OF) =) ‘Again, choosing 0,1,~1 as multipliers, eagh-fidotion of (1) ___detdytde det vez) ODEO Ueryr2) d(x-y) _d(y-z) _ d(x+y+z) Ae=y)y=2) Dat y+2) Taking the frst two fractions of (5), we have dle-y) _dy-2) xy ye (2), Gand (4) => Integrating, los(-y) lol y-2) +s or tog{(s-y)/(y~2)}=loge, or (x=y)/(¥-2)=4 ‘Taking the first and the third fractions of (5), we have 2de-y) , dlety+2)_4 (ey) xt yee Integrating, log(x-) +log(x+y+2) =loge Or (x-y)' (xt y+2)=6 From (6) and (7), the required general solution is o4(x-y)? (x+y +2).(x-y)/(y-2)]=0, @ being an arbitrary function "@ PHAI, i Fr Sera Haar Ka Near LT New D016 Ps (11/RESN, CH: 99104 & 90617, ASRBATIS ema iafioncedeny om: Wale: wre coe‘Cauchy Problém For First Order Quasi-Linear PDE or Integral Surfaces Passing Through A Given Curve In the last section we obteined general integral Pp'+q=R. We'shall now | present two methods of using such a general solution for getting in the integral surface which passes through a given curve. Let Pp#0g=R (I) be the given equation. Let its auxiliary equations give the following two independent solutions. u(x,y.2)=q and ¥(x%y%,2)=¢ (2) Suppose we wish to obtain the integral surface whi 3s tprough the curve whose equation in parametric form is given by Ss x=x(t)y=9(0).2=2(0) oS Where « is a parameter. Then (2) meee Hfa6.9(0.0)) 24. oLe.00 H]NE We eliminate single parameter ¢ from the equatign 0 involving ¢, and cp. Ae Finally, we replace ¢, and ¢) with Sas obtain the required integral surface, . this has form iis envelope surface (2y-2) -4( +? ‘The desired solution is given by the function 2 ‘edited Gy equation(7) Find a complete, singular-~andgeneral integrals of, =@. KN Solution: Here the given equation is (9927.9) =(p? +4?) 9-92 =0. SS O Charpit’s auxiliary equations are ay Putting tis value of g is (3), we get ‘Now putting these values of p and q in dz = pdx + qdy, we have a@=LiP-ey LIF ay op rea a yd) 9 dy NREL ee) Integrating, (z*—a"y')?-=ar+b or 2? ~a"y* =(ar+0) Using @), () gives ay =42 or = Which isa required complete integral, a,b being arbitrary constants,