b . ee 4 vl Sivi Race Sram aes Dedicated to Lord Krishna ‘Authors & PublishesrN an About the Book The Salient Features of this Book are: The matter has been obtained in a simple and eloquent language, so that students themselves shall be able to understand the solutions of the problems. Each chapter contains necessary definitions and complete proofs of the standard results and theorems. These in turn are followed by solved examples which have been classified in various types and methods. This classification will help the students to revise the subject matter at the time of examination without losing any confidence. Care has been taken not to omit important steps so that the students can understand everything without the guidance of a teacher. Furthermore, a set of unsolved exercises is given in each chapter to instil confidence in the students. In view of these special features, it is sincerely hoped that the book will surely serve its purpose. (ir)a ‘This book on “Integral equations and boundary value problems” has been rriculum for MA/M.Sc. students Preface specially written as per latest UGC model eu of all Indian universities/institutions. The text has been written with the idea that thisis their exposure to the subject of mathematics To facilitate the students in their learning process, the text contain some features that may be mentioned here. Each chapter begins with the basic concept and ends with the questions. Further the answers of the exercise are also provided. It is hoped that the students will find these features helpful not only in developing a genuine interest in the subject but also in acquiring the desired level of proficiency in it. We gratefully acknowledge our indebtedness to the publishers and printers for their full co-operations in bringing out the book in the present nice form Ms.Urvashi and Mr. Chaman Singh deserve special attention. Suggestions and comments from all quarters will be appreciated and dully acknowledged. S.R. Singh E-Mail; shivrajpundir@yahoo.com @)ww : Brief Contents Chapter-1: Basic Concept (1-30) Chapter-2: Solution of Integral Equations... (31-82) Chapter-3: Fredholm Integral Equations. ++++(83-146) Chapter-4: Hilbert Schmidt Theory... (147-202) Chapter-5: Application of Integral Equations... (203-258) Chapter-6: Singular Integral Equations ++(259-314) Chapter-7: Integral Transform Methods. (315-340) Chapter-8: Modified Green's Function. (341-360) Chapter-9: Higher Dimensional Green's Function...........(361-402) Chapter-10: Dirac Delta Function.... ++(403-426) Chapter-11: Perturbation Theory (427-460) i)Chapter-1: Basic Concepts... Ll 1.2 13 Chapter-2: Solution of Integral Equations... 21 22 23 24 25 2.6 27 28 Detail Contents * (1-30) Integral Equation 02 Differentiation of a Function Under an Integral Sign 06 Relation Between Differential and Integral Equations 07 + Illustrative Examples 09 ¢ Exercise-1.1 14 © = Exercise-1.2 19 @ = Answers-1.2 19 ..(31-82) Solution of Nonhomogeneous Volterra's Integral Equation of Second kind by the Method of Successive Substitution: 31 Solution of Non-homogeneous Volterra's Integral Equation of Second Kind by the Method of Successive Approximation 33 + Fthustrative Examples 35 Determination of Some Resolvent Kernels 40 © Exercise-2.1 57 © Answers-2.1 57 Volterra Integral Equation of the First Kind 58 Solution of the Fredholm Integral Equation by the Method of Successive Substitutions 59 Iterated Kernels 61 Solution of the Fredholm Integral Equation by the Method of Successive Approximation 64 Reciprocal Functions 67 (vii)2.9 Volterra's Solution of Fredholm's Equation 67 4 Exercise-2.2 81 4 Answers-2.2 82 Chapter-3: Fredholm Integral Equations. 83-146) 3.1 Fredholm First Theorem 83 Prove that the solution 89 Every Zero of Fredholm Function D(X) is Pole of the Resolvent Kemel 91 Ifa Real Kemel K(x,8)has a Complex Eigen Value 2g =H + iv, then it Also Contains the Conjugate Eigen Value to Xo =p —i» 92 3.5 Hadamard’s Lemma 98 35.1 Hadamard’s Theorem 95 3.6 Convergence Proof 96 3.7. Fredholm Second Theorem 98 3.8 Fredholm’s Associated Equation 104 3.9 Characteristic Solutions 106 3.10 Fredholm's Third Theorem 106 4 SHlustrative Examples 108 3.11 Solution of the Homogeneous Integral Equation 119 3.12 IfD(%p)=O.and D(x,8;%9) #0, then fora Proper Choice of Ey, (x) = D(x,8p:2o) 18 s Solution of the Homogeneous Integral Equation 120 ental Functions 121 egral Equations with Degenerate Kernels 4 Exercise 3.1 144 4 Answers 3.1 146 Chapter-4: Hilbert Schmidt Theory...... = (147-202) 4.1 All erated Kernels of a Symmetric Kernel are also Symmetr 148 42 Orthogonality 149 4.3. Orthogonality of Fundamental Functions 149 44. Eigen Values of Symmetric Keel are Real 150 45. Real Characteristic Constants 150 4.6 Expansion of a Symmetric Kernel in Eigen Functions 152 (vit)4.7 Symmetric Kemels with a Finite Number of Eigen Values 153 4.8 Symmetric Kemels with a Finite Eigen Values Roy hy 42 155 4.9. Sequence ofthe pth Power of the Eigen Values ofthe Iterated Kernel 156 4.10 Fourier Series of Power of the Eigen Values of the Iterated Kernel 158 4.11 Hilbert-Schmide Theorem 159 4.12 The inequalities of Schwarz and Minkowski 161 4.12.1. Schwarz Inequality 161 4.12.2. Minkowski inequality 162 4.13 Hilbert's Theorem 162 4.14 Complete Normalized Orthogonal System of CharacteristicFunctions 164 4.15 Coefficients of the Continuous Function f(x) 167 4.16 Complete Normalized Orthogonal System of Fundamental Funetions 168 4.17 Bestel Inequality 171 4.18, Riese-Fischer Theorem 174 4.19 Representation by a linear Combination of the Characteristic Functions 176 4.20 Schmidt's Solution of the Non-Homogeneous Integral Equation 177 4.21 Solution of the Fredholm Integral Equation of first Kind 179 4 Mlustrative Examples 180 4 Exercise-4.1 © Answers-4.1 Chapter-5: Application of Integral Equa (203-258) S.A Introduction 203 5.2 Initial Value Problem 203 5.3 Boundary Value Problems 205 5.4 Deformation of a Rod 207 55 Determination of Periodic Solutions 209 5.6 Green's Function 212 Construction of Green's Function 213, Particular Case 216 5.9 Influence Function 220 4 Sttustrative Examples 222 @5.10 Construction of Green's Function when the Boundary Value Problem Contains a Parameter 244 5.11 Longitudinal Vibrations of a Rod 257 4 Exercise-5.1 258 Chapter-6: Singular Integral Equat (259-314) 6.1 Introduction 259 6.2. Abel Integral Equation 261 63. Particular Case 262 64 Weakly Singular Kernel 265 65° Keer n of the Singular Equation 66 Fredholm Operator 269 6.7 Equivalence of the Fredholm Integral Equation and the Iterated Equation 270 6.8 Prove that the Eigenvalues hy and ty of the Kemelsk and ky are ofthe Same Rank 69. IfaNumbery isan Eigen flue of the Iterated Kernel ky (x, 6), then atleast one of the Distinct Numbers 276 6.10 Integral Equation in an Infinite Interval 277 6.11 Cauchy Principal Integral 278 6.12 Cauchy Type Integral 281 6.13 Cauchy Integral on the Path of Integration 282 6.14 Plemelj Formulae 284 6.15 The Plemel -Privalov Theorem 288 6.16 Poincare'-Bertrand Transformation Formula for Iterated Singular Integrals 292 6.17 Application of the Calculus of Residues 294 + Stlustrative Examples 296 6.18 Hilbert Kernel 300 6.19 Solution of the Cauchy-type Singular Integral Equation 306 4 Bxercise-6.1 314 Chapter-7: Integral Transform Methods. (315-340) 7.1 Introduction 315 7.2. Laplace Transform a15 7.3 Properties of the Laplace Transform 316 ®7.4 Application to Volterra Integral Equation 317 + Illustrative Examples 319 7.5 Fourier Transform 333, 7.6 Application of Fourier Transform, 335 7.7 Mellin Transform 338 4 Exercise-7.1 340 Chapter-8: Modified Green's Function.......:s:sssesessesnses(34 1-360) 8.1 Introduction 341 8.2 Modified Green's Function 342 8.3 Properties of Modified Green’s Function, 346 + Tustrative Excanples 348 8.4 Reciprocity Relation 352 4 Exercise-8.1 360 $+ Answers-8.1 360 Chapter-9: Higher Dimensional Green's Function. (361-402) 9.1 Higher Dimensional Green's Function 361 9.2 Construction of Green's Function for Dirichlet Problems 364 9.3 Conformal Mapping 369 9.4 Table of Conformal Mapping 370 4 Ilustrative Examples 386 $ Exercise-9.1 400 Chapter-10: Dirac Delta Function... 403-426) 10.1. Dirac Delta Function 403 10.2 Sampling Property of Dirac Delta 407 10.3 Derivatives of the Delta Function 407 10.4. Integration of Delta Function 408 10.5. Unit Step Function 10.6. Derivative of Unit Step Fus aul 10.7 Some Properties of the Dirac Delta Function 412 10.8 Three dimensional 8-Function 41s $ Sthustrative Examples 45; 4 Exercise-10.1 423 4 Answers-10.1 425 o)Chapter-11: Perturbation Theory....... (427-460) 11.1 Introduction 427 11.2. Perturbation Orders a7 11.3. First Order Non-Singular Perturbation Theory 28 11.4 Example of Second-order Singular Perturbation Theory 430 11.5. Fist Order Perturbation toan Harmonic One Dimensional Oscillator 431 11.6 Time Independent Perturbation Theory 432 11,7 Time Dependent Perturbation Theory 435 11.8 Eigen Value and Eigen Function of a Perturbed Self Adjoint Operator 439 + Illustrative Examples at 4 Exercise-11.1 448 11.9 Green's Function For Initial Value Problem 451 11.10 Working Method to Construct Green's Function of Initial Value Problem — 453 1.11 Eigen Function Expansions and Green's Function 458 11.12 Application of Integral Equations 461 i)a \ Theory | Ilustrative Examples | Exercise(s) ON Chapter 4 Basic Concepts ¢ S . Z I integral equations form one of the most useful techniques in many branches of pure analysis, such as the theories of functional analysis and Stochastic process. It is one of the most important branches of mathematical analysis, particularly on account of its importance in boundary value problems in the theory of ordinary and partial differential equations. Integral equations occur in many fields of mechanics and mathematical physics. They are also related with the problems in mechanical vibrations, theory of analytic functions, orthogonal systems, theory of quadratic forms of infinitely many variables. Integral equations arise in several problems of science and technology and may be obtained directly from physical problems, ¢¢., radiation transfer problems and neutron diffusion problem etc. They also arise as representation formulae for the solution of differential equation. The differential equations can be replaced by an integral equation with the help of initial and boundary conditions. As such, each solution on the integral equation automatically satisfies these boundary conditions. Integral equations have been encountered in mathematics for a number of years, originally in the theory of Fourier integrals. One of the first results which can be connected to integral equations were the Fourier inversion formulae: w= 25 u(&) cos & x dé wa) ua= 2 I 0@ cos&xd& (2) It can be regarded that the relation (2) gives a solution of the integral equation (1), where u(1) is an unknown function and @ (x) is a known function.2 Integral Equations In 1826, another integral equation was obtained by Abel " A material point under the action of force of gravity moves tical plane (6,1) along some smooth curve. It is required to define the curve such that material point, starting from rest at a point P(x,y) of the curve with 6 ordinate y, reaches the point QG.m in the t any instant ¢, The absolute val Fig Lt f velocity v of the moving point is 250-7] (Bao F]sino Paty=msind By integrating between the limits from 0 to y, we have 2 aenn 0 yaa where (nd =(sino! FY) = [7 (yn! olan, Fl) =0, known as the Abel's integral equation, where F(y) is a given continuous function and 6 (1) is the unknown function. The time of descent can be computed, if the path is known. The actual development of the theory of integral equation began only the end of Nineteenth century due to the works of the Italian mathematician V. Volterra (1896), and principally to the year 1900, published his work on the method of solution for the Dirichlet problem, FEY integral Equation An integral equation is an eq) which the Swedish mathematician I, Fredholm ion in which an unknown function, to be determined, appears under one ot more integral signs, Ifthe derivatives of the function are involved, it is called an integro differential equation. AAn integral equation is called linear if only linear operations are performed in it upon the known functions, Le, the equation in which no non-linear functions of the tunknov function are involved eg. o¢s)= flay Af? Kaa, a Ce) = ASP KCBS ) is a linear integral equation.Basic Concepts 3 and O(a) =2 [2 KEIO) Pas 3) is a nonlinear integral equation Here the function 6 (x) in the equation (1), (2) and (3) is the unknown function while all the other functions are known, These fu the real variables x and & jons may be complex value functions of An equation of the form a(x) 9(2)+ FU)+AJ, Klas) 0) AE =0 is called the linear integral equation, where @ (x) is the unknown function; a(x), F(x) and the kemel of the integral equation K (r,£) are known function ; 2. is a non-zero real for complex parameter, and the integration extends over the domain @ of the auxiliary variable Integral equations, which are lin ar, involve the integral operator Lf KCB) having the kemel K (x,8). It satisfies the linearity condition LLC. (6) + Coo (8)] = GL (8)] + Co L102 (8)1, where L{(&)] =f, Klas) 0(6)48 and G,,C, are constants. Linear integral equat ns are classified into two basic types: 1. Volterra Integral Equation: An integral equation is said to be a Volterra integral equation if the upper limit of integration is a variable, ¢ (290) = Fp s2f KCB) 40M (When siege ep id wena eb fut ht F(a) =f K(s8) 06) a a>—m the equation involves the unknown function @ appears only under the is called the Volteras integral equation of fist kind. (i) Whena_ =, the equation involves the unknown function oth inside as wells ouside the integral sign, then 20) = FA fF Kb) OME is calle the Volterra integral equations of second kind (ii) When 1F(2)=0, the equation reduced to O() =A I? Kn) dE {s called the homogeneous Volterra's integral equation of second kind.a Integral Equations I Fredholm Integral Equation: An integral equation is said to be Fredholm integral equation if the domain of integration (is fixed, eg. a(x) o(2)= Faded J? Kia8) OEM “ the equation involves the unknown function ¢only under the integral sgn, then Fx) =f K(x,8) 6G) d8asxsh is called the Fredholm integral equation of first kind, Gi) When. the integr the equation involves the unknown function @both inside as well as outside sign, then ’ 000) = Fey +A? Kink) 0(@)dE, ash Ws called the nor- homogeneous Fredholm Integral equation of sedond kind (ii), When = LF(2)=0,the equation reduced to 00) =A? Kd) O16) dBase sh called asthe homogeneous Fredholm integral equation of second kind Linear integral equations ofthe fst and second kind ae the partcular cases of the linear integral equation of the third kind, (which are of the form I and II}, where a is not a constant but is a prescribed function of xt a) 9(4) = FO) + A KCB) OCD MB IIL, Singular Integral Equation: When one or both limits of integration become infinite or the kemel becomes infinite at one or more points within the range of integration, the integral equation is called singular integral equation. For example 002) = fla) +A J expl-la-81) 6AE and fa)= 7 @- 87 9G), Ocal are called the singular i integral equations for a fegral equation. The second equation represents the Abel's 1/2 IV. Non-linear Integral Equation: If the unknown function appears under sign to a power n(ar>1), then the equation is said to be a non-lin For example (= Flys af” Kab) OE DME V. Convolution Integral: Ifthe kernel K(s,8) of the integral equation is defined as a function ofthe difference (x=), ke,Basic Concepts 5 K(x,8) = K(x-8), where KCis a certain function of one variable, then the integral equation O18) = Flay Af Ke BOCES a) and the corresponding fredholm equation (3) = Fla) + AFP RC BOCES 2) alled integral equation of the convolution type ‘The function defined by the integral [KH aus =f KOO i) is called the Convolution or the Faltung of the two functions K and @ and are known as the convolution integrals, In general, the convolution integrals may be defined as KG Has (ay Integral Equation Linear® Nomlineat™™ Singulat Voltera, Fredholm (upper limit (domain of integration i fixed) of integration isa variable) First kind Second kind Thad ind [3G ia prescribed iat | fonction of] Nowhemopeneout Homogeneous Fig 12 OG) =FU)+A[, Kin8)ODAE OA)=AJ, Kin) 6 Cae 0 eda (a) + Fa) +2] K (a8) OME aus) 9(2)+ FU+ Af, KOR) OEME =0 ‘el2) 9(0) + Fa) +2] ~ expt-Le- 8) 96) A =06 Integral Equations Consider the function J,(x) defined by the relation, ferentiation of a Function Under an Integral Sign Tyla) =f (2-9 fodan, ) where nis a positive integral and a is a constant. We know that 4 (oe Fur.ndi aed pq) FD 0 8c wide vicinGuni aan ae (Flan ban + Fl Qa] 2 Fs, PON which is valid if F and aF/ar are continuous functions of both x,n and the first derivative of P(x) and Q(x) are continuous. Differentiating (1) under the integral sign, we have Be inf rn)? flodint xm flO pax at, Fe =0-Dlya nL (2) From the relation (1), we have Xe nao = fer finan = Fh = fla) 3) J Differentiating (2) successively m times, we have (n= 1)(#=2) 93) MMV ye a> =(n-D! fla) “ay Thus, we have In general, we have 10) == D1fE [Ef 2 IE Fei tae 8) From thereto (1) and (5), we haveBasle Concepts 7 [ESD LE aid pile ye ain This may be represented as the result of integrating the function f from a to x and then integrating (n—1) times, we J povanta p29 finan (6) HE} Relation Between Differential and Integral Equations There is a fundamental relationship between integral equations and ordinary and partial differential equations with given initial values. Consider the differential equation of nth order as dy a +400 A ay x) y= Fe 1 eras (ty = Fla) “ with continuous coefficients (2) f= 1,2,3,...1% The initial conditions are prescribed as follows WO) = Cy, ¥ (0) =C,, ¥" (0) =O y= (2) where the prime denotes differentiation with respect to 3, a Consider eno) By integrating and using the initial conditions (2), we have aly ox Tet “fg 0OME+ Co dey ox 2 Gok oS, 0OdE +Cuirt Cua : vl y= Jy (Grae +o Sapte ) where | 9(t)atepresents fora mukipl nepal of order n From the relation (3) and (1), we obtain (2) + qf OE ab +ay(f OCG) aer+ ray) * OCG) dB" ° 0 0 =Fa)+ 3 10) “) 1a thee 10) 290) +a aled tot ad ey B04 JF F400) Hap lan(e—8) gla) = Ga) where Gn = Fay+ Sian) Integral Equations (9) yet Spr es =D (6) The equation (6) represents the non-homogeneous Volterra's integral equation of second kind, Particular Case Consider the linear differential equation of second order. @y (x) 462) Ze ay( arta +m) with initial conditions 90) =Cy and ¥ (0) = G, Consider or 4) a and y= Jj 8) 9G) d+ G+ The given differential equation reduces to eyeaer[fj o@as~c] +ata[f 860K] oe (0) + J fala) +m Ns 1 00648 FC)~ Ga (2) ~Gam(2)- Gant) SO)+H JF KC8) OE) 4B, or 4) Where K(x.8)=4 (a) + (8-8, R==1 Fl2) = F(R) Gia (2) C20 (8) - Cyan) ay 2) (3) ay Fa) (5) (6) which represents the Volterra's integral equation of the second kind. Similarly, the boundary value problems in ordinary differential equations lead to equations, Fredholm integralO(2) =sinx +2f costs 806) dé Solution: Substituting the function (x) = xe* in the given integral equation, we have =sinx+2f, Fi cos(x—&).E e8de, [Kanpur 2005, Meerut 2009] =sinx+2 [cos [88 cos B&+sinx fy es sing} =sinx+2]cosx i (cos gsin | Af eS(cos +sin de 7 2 240 as Toles vina| SF cing -ond| “343 Angora] ssina+2| ovr { baet(corxesin ay sin fF et (sin.x-cos oh Foor x{ Feb (con8 +sin§ +sin§ cos of 2 = lo ~}sin x [etoing cos & cos sin of | =sin xt x0" (cos? x + sin” x) — cos x(¢8 sin 8) +sin x(¢8 cos ap =sin x+ xe* —cos x(e* sin x) +sin x(e* cos x-1) =x'=LHS Hence 6 (x) = xe” is a solution of the given integral equation. FERIERA Show that the function 9 (3) =(1+42)? is a solution of the volterra integral equation m= fj eo a "fe cos bed = ping cost bin J sin beds = poplssints ene [Kanpur 2009]Solution: Substituting the function © =1/n Vx in the given equation, we have =o re Je” ays tg a] Thus 9 (x) =1/nVx is a solution of the integral equation. Show that the function @ (x) = 1=.x is a solution of the integral equation fg 8 o@dg== Solution: Substituting the function 6 (x) = I— x in the given equation, we have Jj et Sa-8 ae net fp hater [fe Fae neh (eS) +7 EerS + 08) =eX(l-e*) + 8(xe* +e I) =x =RHS. Thus 6 (x) =1=.x is a solution of the integral equationREEPIEI A Show that the function (x) =e" 22=(273)] is a solution ob thee integral equation. a(n) et 50(G)de =230" A. =2 Solution; Substituting the function 6 (x) = e*[2x-(2 /3)] in the L.H.S. of the equation, ser(2e-J) +28, @(25-F)a = ot(2x-F) +20 (? 3) we(x}+ Blane =RHS. we have Thus 6(x)=e* (2 3) is a solution of the given integral equation Show that @ (x) = cos 2. is a solution of the equation (x) = cos x43 [F K (x8) (EDA, sin x cos &0Sxs& welisie K(%5)= cosxsin & E 5 satisfies the integral equation. wap __y san 0500 e = fsin? @cos = pin? al fa en =| EE Pas ta 0. nso 000 = fxn? Acor Odd Lin? 90 1S1. Verify whether the given function @(x) are solution of the corresponding Volterra's integral equation. (a) O(a) = x= Lota) = [sine BOGE (b) (x) = e%(cose*-e*sine*); 0(2) = (I= xe? )cosl— (1+ x7 )(x-B)9 (6) dB O(a) =5-6r4 J} 5 -6(x-8)] OE) AB——_<—<—<—_$——— ee (BEEPIZEEA Reduce the initial value problem O'-sine@ +e%=x with the conditions 6 (0) = 1,6'(0)=-1 Teg TORchomogencous Volterra's integral equation of the second kind. Conversely, derive the original differential equat Hon with the initial conditions from the integral equation obtained. (Meerut 2004, 2007] Solution: ‘The given differential equation may be written as "(x)= 2-079 (x) +5in rg (x) Integrating with regard to x, both the sides, we have Joe war=fe xdr— Jp e*6 (ahdr+ [2 sin @ (a)de = [ew] = ae [5 © 9(x) de+ [sin x9 (x) JJ cos x9(x) de > o'@)-¢@=32 +sinx9(2)— fF (€% + c0s x) 0(x)de = (x) =-1+ 3? +sin (2) [5 (e* +c082) 9(a)dr Integrating both the sides with regard to x,we have Jo # )ae= fF (14h yde+ [7 sinx 9) te -Sp (e +008 x) 6(x) de = $0) -00)--4 1 4 fF sno eae ~ Jp @-8) (68 + 008 8) (Ea = G0) Ones Lat) +f [sing HUE + cos OOM (1) represents the non-homogeneous Volterra's integral equation of second kind. Converse. Again, differentiating the integral equation (1) with regard to x, we have Oda 4 LJ sing (a -B)(E reo} 008 a8 = O(a)? +f 2 sing HUE + cos slo @ab + [sin x-(x-x) (e* + cos x)] o(x)1 7 eyelet? ff (c8 + cos &) 6(&) dB +sin x(x) (2)Basic Concepts 21 Differentiating with regard to x, both the sides, we have Plas Affi enb glee smerpineansea = e w=rff 2 (E+ con 006 4 (4+ 0089) 9(0)1] +008 x9(x)+sin x G(x) = O° (a) =r -(e" + 008 2) (2) + c05-x9 (2) +5in G(X) = 9" (a)-sin x9! (a) +e%6(a) = which is the required given differential equati (2), we have n. Putting x =0 in the equation (1) and (0) =Land (0) =-1 Reduce the differential equation (2)-36(2) +20(2) = with the conditions $0) =100)=2 into a non-homogeneous Volterra’s integral equation of second kind, Conversely, derive the original differential equation with the initial conditions from the integral equation obtained. {Meerut 2003] Solution: ‘The given differential equation may be written as $" (x) =A sins-2 (2) +39) a) Integrating with regard to x, both the sides, we have J or Gande =a sin xade24* ondrs3 [Fo add = [oe J =-4 (cosj -2f o(adde +3 000015 3 #(8)-9(0)=-A (cos x=1)-2f7 9Oandr+ 36 (0)- Db 3 6 (2) =-2-4(c08x=1)+3[0)-11-2]7 ocree = (3) =-1-4 cos x+34(0)-2 7 (aide 2) - 19) c08 143 90)-2 7 tae Integrating with regard tos, both the sides, we have 2 oad =f aa cos 3 [7 (adds -2 [7 olan? = [ooo] =-+-asine43f% cy adr—2)2 ona? u xndsinx +3 [5 606) dB -2 [5 016) dB?2 Integral Equations = 00) fj 08a -2] (2-8) 04 = ot =x —Asin )+ fj (3-2-8) 06) (3) represents the non-homogeneous Volterra’ integral equation of second king. Converse, Again, differentiating the integral equation (1) with regard tox, we have Tics Tea eat (SS BE BTA Ga) =1-Aeosrt A [5 8-20-9045 = G(x) =-1-4cosr4 ff 315 -20+-9)) 0G) 13-20-29] 002)1 = O(a) =-1-4 cos 1430 (4) -2f 5 4G)aB (a) Differentiating both the sides with regard to x,we : eyo" [ema > ora adsin +39 ()-25 Jy OBE = 6" (2) =4sinx+36(2)-2 6(0) = 9° (2)-3 (0) #26(2) =4sin x which is the required given differential equation. Putting x =0 in the equations (3) and (4), we have 9)=19)=2 BREMIEELA Convert the differential equation a a ee with initial condition y(0) =1, y (0) =0 to an integral equation 2 saan Coster 22-9 wy Integrating with regard to x both the sides, we have (a 4 ip eas, 5 OEM, as ¥'(0)=0 2) ae Integrating again with regard to x both the sides, we have OW = fp 9 ae? = Hs) = y10)= J 016) dB? as y(0)=1 = Way =14 fF =D 4G) @) Substituting the relations (1), (2) and (3) in the given differential equation, we haveBasic Concepts 23 oc-2+ {Jj e)}-afis [J -peerae]-0 = a(x) =3+ Jf Gx-38)0R AG represents a non-homogeneous Volterrs' integral equation of second kind. FEREIEDEY For what value of & the function @(x)=1 2x is a solution of the integral = sa ff toe as Solution: The egral equation is given as Ee 6 a w The function (x) =1+ 4 x satisfies the integral equation (1), then pS +A) = weet feeb a ngyas . : [ae <] a - CF hae AEE La = 1+ a+ Re" het) = recllthtd-e +0) = releirs)-e@4)=0 = 104 D+ eH +1) <0 2 (+ I(rtl—e* = he-l Thus, the function 6 (x) =1+ 2x is a solution of the integral equation for 2 Convert the di equation #6 9,4 2xM_39-0 eae with initial conditions (0) =0,¢'(0) =0 to Volten Conversely, derive the original differential equation with the initial condition from the cegral equation of second kind. integral equation obtained Solution: The given differential equation may be written as eo fe Gras Bese Integrating with regard to x, both sides, we have24 Integral Equations (#y 24h xMdee3 ff oae = ahaa fais Jj oarfed [oat a > Faia Ii - Baa o(a)-2f5 gdre3 [bt = Bros) ode Integrating with regard to x, both the sides, we have (oyp=2 2 xocaer fo on? = (2) =2 ff BoC) dg + fy one? ” $0) 22] 5 So dbs f2 Howe = on) =f + D0G A x @) This determines the non-homogeneous Fredholm integral equation of second kind. Converse. The integral equation may be taken as (2) =fxd+ 2) +f) 1-8) 806) dB - JF (2-8) 806) ..13) Differentiating both the sides with regard to x, we have #0) =fs2n+ J) a- O80 @de-4 J) OE oe 1 a pene eettiael* t6-Benee = (a) =fd42a) #f5 2 tad-E 8006) Af; 24-HE 0G) a = #)= P0424 [ C-OEOOK-f HOR —28 Integral Equations Differentiating again both the sides with regard to x,we have 7 4p a peeea_4 ft teea oO ayals EJ, MBE OG) -F J 50) AE ors. fil anemnpu ree Seman scarD = oa =14 J, 20-90 -[; 2 HoGB-x0(0) - 8a) = 1-290) = $°G)4 (2) 21 6) From the equation (3), we have §(0)=0 and ¢ (I) =I, (6) which is required differential equation together with the boundary conditions (6). Obtain Fredholm integral equation of second boundary value problems nd corresponding to the eo P+ 2g =1,6(0)=0,9'() =0 Gat i = 1010-061 Also, recover the boundary value problem from the integral equation you obtain. Solution; ‘The differential equation may be written as [Meerut 2000, 2005} #&o ooo Integrating, both the sides, with regard to x, we have ad x x jaca fj nde ij otadde = 9(2)- 000) AJy OCB Let ¢'(0)=C, then, we have #)= C+ pP af" 0a Integrating both the sides with regard to.x, we have (3)- (0) = Cr+ 1/6) 39 -AfF o(G ak? = $= Crs (/os? -2f (Que? = (x)= Cr+ (1 /6)x? =A (x8) BdeBasic Concepts 29 Since HW = 099M = Crh Af) AB, = ¢ feaf ods = $6)= 1/2) 240/23 +2 f) ao EAE -A IE (BOE waxes +a{f3 rab! xoeoash HAYS eDOCS = ona aner{f; seeds ! xe@a} oe 1/6)(2° -33) +2) K(x.8) 0(6) dB ) sah King = FAG This Determined the non-homogeneous Fredholm integral equation of second kind. Converse, The integral equation may be taken as or (1/2)x+ (1/623 +245 x9 (6) ALP DOC (2) Differentiating (2) with regard to x, both the sides, we have oe 2 rnd ppmaat fet 1/2) +(/2 +2 fd woe den [5 (xB) ME = o@) C124 0/2)7 +245 OB A] OA.) ating again with regard to x, both the sides, we have araesantitsoreca@a Sterne Martha Jy OGM -AF Jy OGM = wearer], Lo@rae-2f) 26-200 hip (x)50 Integral Equations = #G)+ OG) =x a) From the relations (2) and (3), we have ayes te (0) =O and # W)=-5+5 which is the required differential equation together with the boundary conditions. 000YBE Solution of Nonhomogeneous Volterra's Integral Equation of Second kind by the Method of Successive Substitution Consider the Volterra's integral equation of second kind as (x) = FHA K(x 8) 0G) () where, (i) the kernel K(x,8) #0 is real and continuous in the rectangle R: a Sx Sb,aS€ Sb. Consider] K(x,£)| KG .6) Gad Performing the operation successively for @ (6), we have32 Integral Equations O0)=FU)+ Af? KUO F QRH]? Kort) [> KEE IEG) +f KG. 52)0 G9] dbs = $0)= FHA! KEE FRM +72 2 KCB) KEEFE) ds BFE Ke) JS KG) IS KE 2) 0G) ba dh dB Ingeneral, we have (4) = Fla) +A J" K(x8) FAS +f" KB) REE) FG de +97 KE) [EREED J KEE) 0062) dnd dt HAY Keak) fF KEENE KG Ea) f? KGn2 En) FG 1) pa on yl FHP CEN? KEE) LE KE Eabon fP! KGB) FE) By (2) Consider the infinite series Oa)= Fad4AS Keak) FO +2] Ka af KE, Ey) FG) aad, + PT KLE] E KEES KEE) O62) dbp dbl tat 8) Let Sy) = 2° KG) J KG) IE KG y2-Bea) FG p01 IS @)1=141 [7 KOSS? KEEDI We Since |K(2,8)|sP and] F()|s Q teh 1 Sy(x) | s1a4] Qe" =a" KG p2S na) EG n sla oss-MB lla] a Isyaispq2e=orSolution of Integral Equations 33 If follows that the series is Convergent for all values of 4, P,Q, (b- 4) and hence the series (3) is absolutely and formly convergent. S.C] <}a1) apes Sea) stat appre Ga nly yg (P(b- ayy Salsa Mo Where M is the maximum value of the absolute value of the function 6 (x) which is continuous in the interval I.a<.x x Decos& [Kanpur 2007, 2008, 2010] Solution: (i) We know that Kj(2,8) = K(x,8) By the iterated kernels, we have Kylx8) =f K(ae) Ki (2.8) de = Kye) f Sea fhe Bde te 8) B,x58 se Kind) =f Kerala > Ky(xi8) = ff ef 8(2— Bide = de = Lao D> or ere Sas = HE Fen ade Ele gr. a og = ey de Phus, we have Kyulot)=]f Kone) Kylee (pez seg (2-8 = KyaGs)= [ferret ES yt > Kyle) are - Kale) a8Solution of Integral Equations A Hence the resolvent kernel is given by ROGESA)= FW K (x8) = on RixEsA) = z BEB! ob = ERB) os BYR, (ii) We know that Ki(e8)= Kia) = Oe By the iterated kernels, we have Ky(x,8) =f, K(x,2)K,(z,8)dz > Kalam fe Fe Fea = Kate = fe Feet pees o Ke ff ree ane . We fe- wParam on Kate) [EGE eo BARE ae = Ky@= [2 he-8? Treosg HO? Feet or KC) Jf FASE Ke) de . Keo ]g ee hater ems . Kelas) = Gap eB Fee Hence, the resolvent kernel is determined as RUBE) = YK (8) ro40 Integral Equations or Rnesa)=T A(x-B)" 24 cosx vl D+ cos§ 2408" Au-8) Deeosé Determination of Some Resolvent Kernels (a) Consider that the kemel K(x,8) is a polynomial of degree (71) in & such that it may be expressed in the form K(x.8)=09(2)+ a(t —B) 402) BP. tet (gayle. 1 ee () where the coefficients 5 4,(x) are continuous in the integral (0,4) Let the auxiliary function be 1 t 1 gepmtaast Ga", Sebi) = EGO HALE ROEM de 00) with the conditions ane gel = eno at Hk and atateee 8 # om an a i In addition, we have ean a RUE =T EOE) a) Since the resolvent kernel satisfies the functional equation Rx K Safe K(x,2)R(2,5 54 45) From (4) and (5), we have a : (2.538) = AKU) #2 [2 Keay 6 an (x,8) af ( 46) (5.634) =2 Kiss8) ly ak(az) a2 +n) Key p-MGee @ Using (1) and (3), the relation (7) reduces toSolution of Integral Equations al p De= ot afr wo 2 ita we. vans] 0 (8) The function @(x,;2) is therefore the inte egral of the linear equation Dé =0 which satisfies the Cauchy conditions Thus, we have an expression for the resolvent kemel as R(x,8 3A) = (9) (b) Further, assume that the kernel K(x,£)is a polynomial of degree (n—1) in x such that it may be expressed in the form K,8) = +h OG-1) +O) ata bya) te. FES ec Ee . -.(10) where the coefficients b,() are continuous in the interval [0.4] Consider, Rnbin) = 5 F806 ah) (11) The auxiliary function 6 (2,82) satisfies the following conditions: 1, niger ctateas (12) Therefore the functional relation reduces to a"g e AK R42 fe K(e.8) ota: Adat -(13) Using the expression (11) and (12) and integrating by parts to the integral on RHLS., we have 914 @ 228.29 We]=0 be Sealrott == el (14) Thus, the auxiliary function 9(x,8;A) is the integral of the linear equation D,o=0 which satisfies the Cauchy conditions. Hence the resolvent kernel is of the form ayy2— (15) R(x.E 3A) xan er sa)42 Integral Equations Find the resolvent kernels of integral equation withthe following kemels, a=) (i) Kg) =2x (i) KQ,8)=2-(r-8 (iit) K(,8)=-2+30-8) Solution: (i) Here K(x,8) =2x;A=1 (Comparing with the relation (§ 2.3,1), we have qy(2) = 2, and all other a,(x) = 0 ‘The differential equation (§ 2.3,8) reduces to 9, ae =0;=1 atx=e 41, 2) = = AG (3) From (2) and (3), we have ox Thus, the resolvent kernel is given by la Riaibil) =5 56531 (ii) Here K(x,6) =2~ (4-8): =1 Comparing with the relation (§ 2.3.1), we have @)(x) a(x) =0 1, and all the other Thus, the equation (§ 2.3, 8) reduces to “ with the conditions 9 =O at x =§ 48 tat r=§ 2) Te solution of equation (1) is given by 4 =[AG)+ BG)a}e* @) From (2) and (3), we obtain 4=(8sD Thus, the resolvent kernel is given by La Rew Eros =(e-54 2.5Solution of Integral Equations 43 (ili) Here K(x,8) =-2 4 3(x-8);A=1 Comparing with the relation (§ 2.3,1), we have % (x) =-2,4, (x) =3; and all the other @,(x) =0. The differential equation (§ 2.3,8) reduces to @o 4d ee -30=0 oA) with the conditions o=0atr=§,Salaeag (2) The solution of equation (1) is given by $= Ae + BEe* From (2) and (3), we obtain (3) OnE) = ; et ; 3-8) Hence the resolvent kemel is given by 1é ROE = 7 a O(%§s)= Lat. peep (BRRMTA Solve the Volterra's integral equation (4) = (1-22-49) 4 [°[3 + 6(x—8) 40-8] 6G) ab Solution: Here f(x) =1-2x—4x?;2 K(x,8)=3 + 6(x-8)-4(r- 8? Comparing with the relation (§ 2.3,1), we have 4% (2) =3,44() = 6a (2) =-8 The differential equation (§ 2.3,8) reduces to (1) with the conditions @ = =— =A) The solution of equation (1) is given by 6 (x,8 51) = A(B)e + BE\e2* + Cle (3) From (2) and (3), we obtain (x, 83) 1 -ax-8) , | acre) tig? aa"ry Integral Equations ‘Therefore, the resolvent kemel becomes 3 ninsin= 1 gatsn 1 2-8) _ 394g 8 abe 8 4428) 324 (x2) a) + Thus, the solution of the integral equation is given by Lara? -324-}) [tes eset x(l-26 -48? )dB (Ref. § 2.2, (1) Lr et 22, or 1 [ertosi0g +42) per eag BE?) + cMt9)(-328 -32 or (2) #1-2x-4s? 14 Dede? bel wet Alternate Solution : The integral equation is given as (8) = (2x47) + J 13+ 662-8) AU - B71 06) A (ly Differentiating both the side with regard to x,we have 0) =(2-89 +4 J 8 +601-8)-40-B J 06M = 82-80) f) 215 +6-8)]-40-GF HOG +300) = (2) =(-2 81) + J [6 -8x-8)] 016) +3012) 2) Differentiating, again, both the sides with regard to 2, we have ow 8+) 2 (6-Ma-8)] 9(6)45+6900)+ 34x) = (3) = 8-8) 5 6G) dB +69 (2) +300) (3) Differentiating the integral equation (3) with regard to x, we have 9° (2) =-89(2)+69'(2) +3 9°(0) = 9" (2) 39" (2) -66'(x) +85 (x) =0 (a) Substituting x =0 in (1), (2), (3) and (4), we have 40) =1¢ 0)=+16"@=+1 3) The solution of the di ferential equation (4) is given by OO) Get +e eet 6)Solution of Integral Equations 45 From the conditions (5) and the solution of the differential equation (6), we have $0) =13G4+G+G = 90) =1G +4G -2C, =15G =1G -0= CG 9") =13G +16 +4G=1 Thus, the required solution of the integral equation is given by (2) =e" [EEIMIEEA Solve the linear integral equation : (2) = (cos x—x—2)+ f° E— 2 9(6) ae Solution: Here F(x) = cos x-x-2;4=1and K(x,8)=€-x Comparing with the relation (§ 2.3.,10), we have a®= Thus, the equation reduces to. &’ +0=0 e , and all the other b,() =0 with the conditions we(L) 0-0 at §axand Palas s (2) The solution of equation (1) is given by (x,8 51) = A(a) cos & + B(x) sin& -(3) From (2) and (3), we obtain (x85) =sin(&-x) Hence the resolvent kernel becomes Rey Es) --t Soe. 31) =sin(& - x) AA) Thus, the solution of the integral equation is given by (2) = (cos x—x-2) + J sin -x)[cos § -§ -2]a8, or (2) =cos x- 3-2 +coss[ cos 2 +E cos E-sin § +2 cos f46. Integral Equations wsins[ 8+ jsin25- sing con -2sing] or 02) = cos r= 2-24 cos x[- feos2e4 re0s 2+ 2c08x—H] ~sin xf}sin2r—ssin.x=2sin e+ 1+ ba] 1 or 9 (2) =~cos x-sin x Solve the linear integral equation: (2) =29-+6x+ JF [5-66x-8)] 0(8) a Solution: Here f(x) =29 +6x;) =1 and K(x8) ~6(x-8) From the relation (§ 2.3,1), we get 4) (2) =5,a4(2) =~6;and all the other a,(x) =0 Thus, the equation reduces to a a-Si 69=0, a) with the conditions o-Oatr§ Palate ng 2) The solut of equation (1) is given by (x8: =A HIP? + BOP? (3) From (2) and (3), we have O( 6) 2Ae9 —2e-9 Hence the resolvent kemel becomes ent” gene Regia 7 Fy (ni) or Regi) =920-9 4 20-9) ) Thus, the solution of the integral equation is given by (0) 29 4604 J 920-8 420-9 (29466) 45, or (2) =29-46x4(-878-D 4 5820-9) ~54 1h EUS) SUB) 4 D4 (LEU) pero or (2) =29+ 61-29-61 + 938" — 642* = 938" — 647*Solution of Integral Equations Az (ERIE With the aid of resolvent kernel, find the solution of the following integral equations: @ 6@) =sinx+2 a So (8) de iis #24 cosx Gi) OG) =e* sine fF Teeose OE [Kanpur 2002] Solution; (i) The resolvent kernel of the kernel K(:, 8) =e Sfor i =2 is given by R(x,8;2) = 202-8) Hence the solution of the integral equation is given by Oa) asin 42 J BOS) sin Ede or Ota) ssin e422 1 6 55(3sin8 cos of or (2) = sin x— $69 [343 sin x +-cos x)—I] or OC) =FeO* Leos rt 2sinx (ii) The resolvent kemel of the kernel 2+cosx BP een e for = lis given by Rast) 0 Ee Thus the solution of the integral equation is given by Olam sins fo FOE OE sing 7 9 (2) =e sin r+ 67 +008 x) [fF ret 3 oe (x) sinaaroea[ Fes} = o(x)= es ifA=1 => (2) =e* Gi) 9) =14 fj rE 0G ae Here Sle) = LK (2.8) = 38, Since KB) = Kin8)=3850 Integral Equations aa Kalas) = ff Rea K(b)de = Ky(x8)= fp ree fab(x? -89) = h(t - 264) or K(x of, K(x,2) Ky(z,8) dz - y(n) =[f se ftete-a8tyae - roy heee-2eeh ot) or Ky(2.8) # kt) Kylee = K, (276-2284 2k?) de = Ky(8) Hg 3764 a 3x4Q7 — a8 ) = Tgp e ) and so on. Thus, the solution of the integral equation is given by Ua)+ flad+ DY ff UK ELA = OILED fo KiB [EAB 7 OUada de fo Kyla ab tJ Kaleb de [2 Kel 8)a + * eae fate att ae = OH Le fh bad 8 te a8 as og ty-aate oath ag ect flea telaggute? — lO) any, Tag lo (98-3276! +3288? — 281) a8, a sale» Gel-Gel]Solution of Integral Equations 51 1 Ley" si #3) o x “raf GPL rou dH ey teh, PREEPIMEAN With the aid of the resolvent kernel, find the solution of the integral equation Oe + fe ge ae Solution: Here K,(x,8) = K(x,8) =e? -@ AL) and K,(x,8) =j, K(x,2) K, (2-8) dz (2) Substituting v =234... in the relation (2), we have Kye8) = JF Kena) Ke Bae = Kye) = JEP Bde ae gee (3) or Kg(x,8) ae K(x,2) Ky (2,8) de = Ky(x.8) =f ee 8 (ede oe ae AA) or Ky(a.8)= Jf Kaz) Kyle. Sue ~ Ka= [pee es ooh ge OB ee 5) or K,(x,8) = oar ey 212.3)... Hence R(x BA) = z WEEK, (3,8) = K,(x,8) + AKy(x,8) + 22K (3,8). a = R(xbsa) =e jot oeset + ROxE:A) ae eS The solution of the integral equation is determined as52 Integral Equations a fee Pete = 40 +e = 4c) PETUEMERE Show that 8 (x,8:4) = K(x8)+A [7 K(2)R (28:2) dz isthe solution of 4 non-homogencous Volteras integral equation of second kind o(x)= fiaynr fy K(x,8)9(6) a5, where R(1,8;) is the resolvent kernel of the equation. Soluti m: We know that the resolvent kernel of the integral equation is given by RUngia)= EG «) The iterated kernels are given by K(x,8)= Kind) and Kx8)= J 2 Kez) K, or R(x,E 5A) = Kx) + = =K(x,8)+ > PE eas = R(x,EsA)= K(x,8)+ DM fF Kae) Kyle bees mos > RUx.GsA) = K (6) +2] F Wy (e8)| Keele = RongiA)= Ka) A] ¢ ReSi9K 2) de = Rix §:9)= K (x8) +2 fF K(x2)R (Bade Solve the Voltera integral equation of second kind, by using the method of successive approximation: (060) = 4a)- Jo 6G) a with a(n) IL Jo (#- 840s fi) 906 with @(x) =0Solution of Integral Equations 53 i) C2) =14+ J Ce-8) 0G) a8, with yx) =0 iv) Ca) =14 J 606) dB, with (2) =0 ©) (2) =(143)+ J 8) (8) dB, with (2) =1 Solution: (1) The integral equation is given as 9G) =U+9~[F ods Here F(x) =1+ x, K(x,8) =and 4 =-1 The v th order approximation is given by O(a) = flay + AIS KUBO or Hla) = (14 2)~ J 1 OB Substituting v = 1,2,3,....,we have O1)=O+2)~JF Gade = U4 2) fae =1 2 (x)= (42)- Jo Gab =(+2)-3=1 $3()=(1+2)-[} ® @ae=U4+x)-x=1 (2) =(142)- Jp 1 G)dB =1 Hence the solution of the integral equation is given by 6 (x)= lim 9,(x)=1 (ii) The integral equation is given as $=2-Jp 8) OO) a to(x) =0 Here f(x) =x, K(x,) =x-E,and A =-1 The v th order approximation is given by HD=fNFALS KEHOE34 Integral Equations = 0) = Jf 8G) 4B Substituting» =123,...,We have Ux) =4- Jj =e Oa =x [f0s-2yae-x-8 etn) = Jp 94 a= eG) [5 6-DO OE ” oC) “as a +(-1)' @r=p! . serene (iii) The integral equation is given as Jp (8) 9@)as with @ (x) =0 OG) Here f(x) =1, K(x,8) =x- Band =1 The v th order approximation is given by 2) = fA) +A J KEE) G1 GAB = HD=1+ ff 8) AB, Substituting » =1,2,3,.... we have G2) 14 J. 2-8) oy Oe . a 2 or Go) a+ 2 6-84 Oi als [5-9 ale or tO) =14 [50-9 OaSolution of Integral Equations _ x 2 (x)= 14 f2 eal S la aif .8 8) 2 = 3 (x) us 38 a eo aT al In general, we have 2 2 Ba Hence the solution of the integral equation is given by (2) = lim (2) 4 > $a), + => (x) =cosh x. (iv) The integral equation is given as (3) =1+ J @).d6 with @ (x) =0 Here SQ) =LK(x,8)=LA=1 The v th order approximation is given by O02) = FA) +A IE KE) 0,4 as = Orla) =14 JF a Bae Substituting = 1,2,3,.....,we have A) =14 JF Bak =1 or (2) =1+ J Ode = 14 ff dg = 1+ x s 2 or Ua) = 1+ Jy Bae a1 J) (1+ Bas = 1+ 4 In general, we have 3 vel eGnieee= 4% = nate Hence the solution of the integral equation is given by 6(x) = lim 6 (2)36 Integral Equations - oajeleses (© The integral equation is given as 2) = 4294 JF eB 0G) Ag (2) =1 Here SG) 14. Ke) nd ‘The v th order approximation is given by La) = f+ A JS KCB) G4 Ae or (2)=(42)+ Jp 2-941 Gas, Substituting » = 1,2,3,....,we have QU) = U42)4 fF Bm Bas 4 areas [fo-9a * toed=densfZ enon ~ torneo [50-8 [tes “ ew-dens{(s- ar Z - wivreran(2-2),(2 Pos iw “ perio 2 a Ingenenwehive (alc +f Hence the solution of the integral equation is given by lim (2) =1424 3 40—_— 1. Find the resolvent kemels of the Volterra integral equation with the following kernels: (a) K(x,8)=x-& (b) K(x8)=0° @ ©) K(x8) = SF (@) K(n8)=c8c>0 2. With the aid of the resolvent kernel, find the solution of the integral equation (@) o(x)=e8 + ff FoR ae () o()=x3*- J) 3° Fo) a () o(x) mer Pale or Sede (4) 6(4)= fla)+ Jj e508) dB INSWERS-2.1 — (a) psin hyk (x-€),2>0 (by eA(anB)_a? 8 (©) coshs nung (dS eA) cosh & (a) $(2)=3"(-e™*) ) o()=2* © o¢a)=e"*0+24) @ oy= fet ff A fede38 Integral Equations pF 9 Volterra Integral Equation of the First Kind Theorem, Volterra integral equation of the first kind can be converted to a non- homogeneous Volterra integral equation of the second kind An equation of the type JZ eB 0@ee= feo “ is called a Vol a integral equation of the first kind where K(x,8) and f(x) are given function and @(x) is the unknown function, Assume that the kernel K(x,6) and all its partial der tives © K(x), flab fla) and f° be att ‘nd <0 nay een tlhe» Monae, the kame Kio) poset» conums date 2K he fit member of the equation (1) also possesses a continuous derivative, the same must then be true for f(x), Differentiating both sides of equation (1) by Leibnite's rule of differentiation with ri d to x,we ol K(x.2)0(0)+ J} Ks 8) OO B= LO) 2) Conversely, eve solution of the equation (2) also satisfies equation (1), since the two members vanish for x =O, and their derivatives are identical If K(x,x) does not vanish at any point in an interval [0,4] then equation (2) may be fis) px Ky ) Key fe Renny 2OS 8) 4) which reduces to a Volterra integral equation of the second kind. If Kes which we can treat in the same way, provided that K(w,&) possesses a continuous second then the equation (2) again reduces to an equation of the first kind, derivative, Again, differentiating (2) both the sides, with regard to x, we have Ky'(nis)@(0)+ ff Keb) 006) dB = f"(8) (4) which represents a Volterra integral equation of the second kind provided.Solution of Integral Equations 59 Ke(4.2) #0. 1F Ky'(x,2)=0 we adopt the same procedure again, and so on. Thus, we have the sequence of successive derivatives with regard to x of the kernel K(x,8) until we get a derivative K’"l(x,8)40 for x=£ In onder that the equation ()has a continuous solution, it would be necessary for f(x) to possess continuous derivatives LDL)... f(a) which are all zero for x =0. The first (v1) equations obtained by differentiating (1), both the sides, are satisfied for x =O, Ke) 60)+ fF 2 Ke. De@ ds =P) =f") * Ke) = PON RG bo ay. MOREA which is an equation of the second kind provided K(x.) 40. 2M Solution of the Fredholm Integral E quation by the Method of Successive Substitutions Consider the Fredholm ‘integral equation of second kind as b (x) = Fla) + Af” K(x.) 6G) a8, AL) where, @ the kernel K(x, §) #0 is real and continuous in the rectangle R:aSx R(x EA) = YK, (x8) AM Kat The series on the R.H.S. is called the Neumann series of the kernel K(x,&).If M is an upper limit of | K(x,8)| then the iterated kemel | K,(x,)| < M"(b-a)"" at every point of the domain D. The series (7) is therefore uniformly convergent in this region, if the Parameter |A|<1/M(b—a). Thus, the relation (2) which gives the solution of the integral equation (1) assumes the form (0) = Fa) + Af? RULER) Fae (8) The function R(x,8;A) is called the resolvent or the resolving kernel for the equation (1). The equation (8) represents an explicit solution of the integral equation in terms of the resolvent. In view of the relation (7), it follows that the resolvent kernel R(x,&;4) satisfies the functional equation: R(x,8:4)— K(x, 8) =2 YW Ky (a,8) P=) Rx EA) — K(x,8) =A K(x,2) DU Kya (8) dz r= R(x.E iA) K(x.) = Af K(x2)R (2,650) de REA) = KB) +2 [Key 80) de66 Integral Equations In view of the relation (7), it follows that the resolvent kemel R(x,§ 2) satisfies the functional equation: AY UK 28) Rex B:A)-K(n8 J K(2)S 2K n(e.8)de r= R(x, i)~ K(as8) Af Kase) REE ADA R(x8:2)- Kin) = Reba sh) de Keg) +2 J? Ka2R Ren Bia) = King) +2 [KER (ide 0) The first equation is established by transposing x by z in (7), then multiplying both sides by K(x,2) and integrating term by term in the interval (ab). Similarly, we may: Now, if the resolvent kernel R(x,8:2) satisfies the equation (9), then the function 6 (x) in the equation (8) satisfies Fredholm integral equation (1). Substituting 6(x) from (8) and (1), we have F)+a [? ROLE FE) d= Foytnf? K+ {FO ‘ +a! Rein) Fea By transposing & and z,we get Ftd [2 RG.E:2) FR) dB = Fanta? {KO sal? K¢ Reeve FM (10) From(9) and (10), we see that the functional relation (8) is the unique solution of the Fredholm integral equation (1) when the absolute value of the parameter 2. is sufficiently smallSolution of Integral Equations 67 Ea Reciprocal Functions If K(x,6) is real and continuous in R then 3a reciprocal function K(x,£) provided that P(b~a) <1, where P is the maximum of| K(x,8)| in R. We know that the iterated kemels of K(x,£) are represented as Kyl8)= J? Ky(0.2)Kym (8) desm cn and K,(x,8) = K(x,8) Let ~K(x,8) = K(x,8) + Ka (x8)... + Ky(a,8) + (1) When K(z,&) is real and continuous in R, the infinite series for uniformly convergent if P(b-a) <1, Thus, we have WHE) — KC, 8) = Ky (2,8) + Kg (x,8)4..4 Ky(.8) +... “MB) K(2,8) =|" Kx2)Ki 8) de (x,8) is absolutely and or wtf? Kz) Ky(2,8) de +... or MB) K(2.8) =f? Kone) Kea )de +. +. ot JE Reale (e dee or ~H(x,8)- K(x6) =? UK (2) +t Kn} KB) dz From (1), we have K(x8)+H(2,8)=[? K(x,2) Me,8) de or K(x.8)+ k(a8)= J? kez) K(e,8) de (2) The functions K(x,&) and K(x,8) are said to be reciprocal if they are both real and continuous in R and satisfies the condition (2). A function k(x.) reciprocal to K(x,£) will exist, provided the series (1) converges uniformly. Volterra's Solution of Fredholm's Equation A function k(x,8) reciprocal to K(x,8) exists where K(x,8) is real and continuous in R,K(x,8) & 0, f(x) is real and continuous in J, f(x) ¥0, then the Fredholm’s integral equation of second kind has one and only one continuous solution in the interval Lasxsh.oe Integral Equations Consider the Fredholm integral equation of second kind as, ’ o= fers J Kix 0Oe, “) where the reciprocal function (1,8) of K(1,6) is known, Let the eq) continuous solution @(x),then 8) = rerrf? KEE) (Ey) aby Multiplying by &(x,) and integrating over the interval (a8), we have . 6 4p JipHeB) 06145 = JP RCE) FEDAE +f F PACE) HEE) OE) Aas - Jaca) 0065 =f? Ha. 8) DAB #f? | KC) + Ha) | 006) = om fe esteae + f° (6) 64, Again [°K (3.8) 006 Me =9(0)- fl) = 90) = fx]! Ken8) /E @) It follows that the integral equation (1) has a continuous solution and is given by the equation (2) which is unique MEITIEEER Find the iterated kernels for the following kernels () K@8)=2-§ if a=0,b=1 (i) K(x,8) =sin(-28):0 (ii) We know that K\(n8)= Kin) = a Kyatyfi or K(x.) = Ke(x.8) = In general, we have Thus, the resolvent kernel is given by RUB:a) =F AH KC) =e 0 Solve the following linear integral equations © by=ref” oO pe? erga glo 58945 i) (x) =(sinx—2) + (i) 9) =Ginx—D) Solution: (i) Here K(x,8)=1, =1 and F(x) =x Weknow that — Kj(x,8) K(x.) or Ky.) = fq" aSolution of Integral Equations 73 or Kyat) ay" Thus, the resolvent kemel is given by Renkin aS ay” =2 2-4 or R(x,E;1) =2, asd=1 Hence, the solution of the integral equation is given by OC) =F) AS” RULER) FEA 12 1 or o=+]) 26 d6 = x4 2 = 2x4 constant (i) Here K(.8)=82,2.= 1, Fla) =sine—* We know that K,(x,8) = K(x,£) = && or Ky (x8) = 385°? a2 Bde : a = Ky(x.8) = 28 f°? a=3(5) x8 G 7 Ky(esdh=fP" ant (3) ae = Ka (x,8) = i J [07 Pa or K, (2.8) = mie ae Thus, the resolvent kernel is given by 4 ROEM Tae Ba x or R («85 96 =o 96-1a” Integral Equations Hence the solution of the integral equation is given by x, 196 4s (0) wsins 24 TPO 7 x, 24x 96-n3_ or (0) =sin E+ PAE EE asin EREIIEEZA Soive the integral equation Soe) [Kanpur 2005] Solution: Here K(x) Weknow that — Kj(x§)=K(x18)=§ 1 or Kage fy esas or Kylub= 5 fg 8dem a8 & 1 or (a) Hence, the resolvent kernel is given by nwened 2)” 4 1 oe Rowsif) «4 a2 Thus, the solution of the integral equ ion is given by Ba de 12 (3 a 1 wer peget zh 8-3 « teo-(Q4-Le 2)Solution of Integral Equations 75 [2 Bh Les 288 +28) 8] 3 1 1 or sie lyr $@)=5 pe gett PREMEIA Solve the following integral equations @ 0) = 24h ft XE O(8) dG Gi) Ge9=(eF-Ferd a}tFh, 6a Solution: (i) Here K(x,£) = x8, =} F(x) = Ps Weknow that Kj(x,8) = K(x,8)=28 or Ky(n8)= J) seth = hag a es Kng)=[) 23 %-(3) x or Ky.8) “Q's Hence, the resolvent kemel is given by md S (AY _ 3x8 R@EM=nED (S] = (3) <3 or R(x,83}) = $ x8, Thus, the solution of the integral equation is eo by 5x16 bei $0) 42 Sl 85 oH 66 (ii) Here K(x,8) =1LA=},F(x) =e -he+d Weknow that Kj(x,8) = K(x.) oe Ky(n8)= ff &=1 7 Ky(n8)= J) =I76 Integral Equations or Ky(a8) =1 Thus, the resolvent kernel R(x,8;2) is given by Rueg:a)= yt 1a? te or He) aCe fers fh Efex has : a(t a $0) = ~beobs[é-( Solve the integral equations; ) ocd seen fe Foayag (i) O(a) =1+A [sinter +O a, [Meerut 2003] Solutions. (i) Here K,(sx or Ka (x8) Proceeding in this way, we find that all the iterated kernels coincide with K(x,6). Thus, the resolvent kemel becomes P(x,B:A) = K(x E+ A+ 4...) ALA) ‘The resolvent kemel is an analytic function of 2 except at the point 2 =1, which is a simple pole of the kernel P The solution of the equation is given by = fay f! eb fe 6) = fd J Oe (i) Here Kj (3.8) = K(a,8) =sin(a+8) ot Kyla) = [Ssn(x+e)sin(¢+ Bt =f mcosle8) or Ka(as8) =} e [2 sin(x+ t)cos(t—Eyde= (hx? sinters 8) Proceeding in this way, we have K, (1,8) =(G m)? cos x8), Ks (4.8) = (G2)! sin( + 8) Kg(2.8) = (9 cos(x—8),K, Ja) singe + 8)Solution of Integral Equations 77 The resolvent kernel R(x,8;A) i given by REA PK (r,8) = OF R(xEGA) = Kj (x,8) + AK (x, 8) 4:22 K3(x,8) +23 Ky(x,8) HA KS (1,8) +95 KK (2,8)... 2 4 or Resbiayesncees fi (28) +(8)' +. } 2 a + Feote-sfto(*8) “| 2a5._[i, an an) (an)! or Rexb:A)=[sinte + + caste pC) +(zys nls an 1 or Rox8:2)=[sin(a+ D+ Fem 0 | Provided ad staal or RUE = [2sin(x+&)+ An cos(x-&)] Thus, the solution of the integral equation is given by 90) = f+ Af RULED) flO db > O(x)= Maal {2sin(x+ &)+ Am cos(x-E)}dE > o@)al+ 7 a [+2 cos(x + &)—Ansin(x-)]5 => (2) 14 oa gy (2 cosx+ Ansin x] where| |< 2 Solve the integral equation O(a) =142 J) (3x8) 6 ER For what values of & the solution does not exist ? Solution: The integral equation is given as O(a =14 Af) (3x0 Al) Here (x) =1,.K(x,8) = 1-328, = Isay) Weknowthat K(x.) = K(x,8)=1-33% (2)78 Integral Equations and K(x i K(x=2) Ky (2.8) dz (3) Substituting v =2 in (3), we have Kylx8)= J) Kove) Ke Bde & Keak) =f dS) See or Kyla.e) fp, {1-32 +8) +9382} de 3 or Ky) =1- 3 +8) 4398 (a) Substituting » = 3 in (4), we have Kyle, fl se or Kyle, J,.0-39[1 +9+328] pet! [r3e)-9e(!-eny-3t)eou?(1- eae a K(x) = JQ [(-35)-G grr at) +9m0?(f jie or Ky(ni6)=t0-328)= 1 (08) 6) Similarly Kyl8) =f} Kes) Kj (2.8) or Ke. 4p, @=3x2)-328) de or Kyoty= Ffi-5 e438] 6) and Ks (x8) = J) Kox2) Ky(e.8)de or Ks(28)=5f) d-Sefi-F +o sssthae 2 a or Ke.8=(3) c-38)-(G) mes) oy Thus, wehave — Kj(2,8)=1-3x8, 3 Ky(x.8) =1-3 (+8) +338 olay Kunt)=4a-38),Solution of Integral Equations 79 Ky(,8) = ‘| K5(x,6) = ay (1-328), pot oes) oes8d-() f-Zecserssa} The resolvent kernel is given by REA) =F AK, (x8) mA or ROG EIN) =[K (8) + 2? Ky (4,8) 48K (x,8)4...] 1A (4,8) + PKG (1,8) + Ke (x,8)4..] or Rid) 0-328] 14 eB EA) (1 Lae (1-338) 14+ 5 | oft perpesalfied or R(x EA) =| 3x8) 40 —_ Eo are { peroeaal| (74) 4 . 1 or Rubian=— 4, [}+4-Jae-a(< 2-1) (8) Hence, the solution of the integral equation is given as O(a) =f) AS) ROEM O MA or eeyatr rf 5 fia She ae(x+5a- s)he 4a 44,A- 44212 -3. or eo) =1+ 4, ee a <2 (9) From the equation (9), we notice that the solution exists only when |2|<2. The solution of the integral equation does not exist for| &|>2 Solve the non-homogeneous Fredholm integral equations of the second kind, by the method of successive approximations to the third order. (O64) =24+ 24) (x48) 0d G() = Gi) OG) =142 Jf (+ 80GB. OyC2)=180 Integral Equations Solution: The integral equation is given as (a) =2x4 Af) +8) 6G)dS oy) =1 a Consider 4,(x) denotes the nth order approximation, then, we have u(x) = 24+ Af) + 8) 1 DME 2) Substituting m = 12,34,....in the equation (2), we have WO) HDH AG) (r+) Odea IeH A) HEYA or a(a)-2e4a(x+4) 8) Also (2) =2840f) (+O or olay =2a af) tro) frsen (se 5) Las or (a) =244 J) (x48) Gas ees or (0) =Br0n(v0)o72(x4 2) (a) and (2) = 2x42], (x+ 8) CG) or aatsymaren ft twos) fogea(s+2)o2(5+ 2) « woraeefiaen 2 Bassnatela on ne e ty(o)-2e4af! (2 * 7 (3 Mert hee Peottawre he 2), 92(7, 2). 3(!3 65 or te(o) 2a in(xeZ} oi (ZareZ) +08 (15 v8) (Gi) The integral equation is given as $0) =142 J) +9 9OE.G A() Consider 4,(x) denotes the mth order approximation, then, we have ADH TEAS) (+B) ME (2)Solution of Integral Equations Substituting »=1,2,3,4,.. or Also or or or and or or or -.-in the equation (2), we have AEDT ALS (+8) Oab as af! (rs Ea aG)=142 (+43) --@) @O)=140 J) +84 Gade ew=14a)) ces siftea(sed hae e)=14a)) [[s-24) ++ (tga sre)e si %Q)= tealrsp)oit(xe 2) (4) Gla) =14A J e+ BO as Bedatea ly oben (sedan (2 2)Lae aa)= Leaf) {ie} a a} 2 [Fe Feaent}enaenes Qa)al+a (ned) 7? (s+ o)e0 (iSe+3) @rcxcise-2.2 — 1. Find the iterated kernel of the following kernels: (a) K(x§)=sin(x—§);4=0,b = n/2 for n=23 [Kanpur 2002} (b) K(x,8)=2e5 ;a=0,b=1 (©) K(x8)=e3) ; a=0,b=1,forn=2 Construct the resolvent kernels of the following kernels: (a) K(%§)=1 ;a=0,b=1 (b) K(2,8)=sinxcos' (ce) K(x8)=s654 @ Kuxg)=P8;INS WERS" 2.2 (a) (b) (©) Ka (2,8) =} sin(x+8)—4 cos(x-8) Kg(x.8)=4 Ky(.8) = xe 2 =z? .. 16 sin(x-§) 1GEb= per + e?-2-8] 4 (E-x-De 7 0sx 8 2 (455) = L pers + 2-2-8] 4 (x-E-Ne™ Exes 2 h R(x,G50)=2 R(xG sh) = 7S a] <2 Rex gh) = a <32 Z R82) = 2S jaesi2 O(x) =x $@)=2 2123 3-2 (x) =[4+44(2 -32)]/(4-#)e have determined the solution of the Fredholm integral equations as a power WwW series in the parameter A, uniformly convergent for |2| sufficiently small. Fredholm obtained the solution of the integral equation in the general form, if possible, for all values of the parameter 2. In the theory of integral equations the well known theorems of linear algebra which are related to solution of the system of linear algebraic equation play a leading role. Now we shall discuss the solution of the non-homogeneous Fredholm integral equation of second kind by replacing the integral, appearing in the equation with a sum which reduces the equation to a system of linear equations and assuming the number of terms of the sum tends to infinitely. Fredholm First Theorem The non-homogeneous Fredholm integral equation of second kind O()= FDA? Kins ods, under the assumption that the function F(x) and K(x,€) are integrable has a unique solution, is of the form b (x) = Flay Af? R(xE:A) FE ME where the resolvent kernel R is a mesomorphict function of the parameter 2, being the ratio of two entire function of the parameter 2 R(x,§ 32) = D(x,§ 5A) / D(A), D(A) #0 ‘A function f(z) whose singularities in the finite part of the planes are poles is called a meromorphic function.24 Integral Equations defined by Fredholm’s series of the form Daas SB" yh! sree ak ) lg wed Lee cate and Diwbia)= Kons) Yh? fae BEE Ja ate By Spee ‘These series converge for all values of A. In particular, the solution of the homog integral equation is zero, Consider the Fredholm integral equation: + y= Fy 4af? Kia) OG) AB “ with a Riemann integral in a given interval (a). In accordance with Fredholm method, wwe consider the partition of the interval (a,b) into mequal parts by the points = 081 By B5rncen bmi be where Go =a) =a+ hey =a+2h, and hE pay by =b-a) /m 2) Ey =a Replace the definite integral in (1) by the sum, corresponding to the points of division, we have $(2)= Fla) 4S KUaEy) 6G) (3) (2) — NHK (4451) 9 (Er) + KC 89) 6 Gp) torch KB 9) 6 Gy)] = Fla) Since the equation (3) holds for every value of x, it must be satisfied at its » points of division x = & 8) .83)--.-By- Thus we obtain a set of m linear equation with 1 unknown values of the function 9 (E)),6 Gp )e---6 En) 4G )— WAKE, 5) 6 (81) + KE) Ep) (Ea) + +K(G Eq) 6(En)) = FG) £1) (61) + KGa Bo) 0 Ga)+: +K(G.£n) 6G] = Fp), 0G) - IKE,Fredholm Integral Equations 85 9G) MACK Es £1) (G1) + KE. 2) (G2) + +K(Gs.5n)4(6n)] = FAs). Gy) ALK (Gy) (61) + KE 82) (Ba) +K Cubs) 06m FE) = OKIE 6) = Ky The system of equation (4) reduce to (1K) ~My — Kg yoo Mig y= Mins + (1 Mi p9) & ~ WeKng ~My y= By (Gn) (A) With the notations F(E) AIC + Hip Oy + (1 Wig) 5 —— MAK 9 = By DIK, 9+ MiKo by + (= Mi) 4 MK py = Fy (5) The solution @, 2... of the system of equations may be expressed in the form of the ratios of certain determinants by the common characteristic determinant MI-AWK)) “Ky Kg... AUK, ~MiKa, (INK) MAK... -AAK | ay_| ~My ~MAKy (I= 2iKyg)... AK] Da = a a 31 (6) UiKy MK -AiKyg (LIK) provided that D,(A) #0. Now we shall expand the determinant (6) in powers of the factor ~Ait The first term not containing this factor is obviously equal to unity. The term containing (Ri) in the first power is the sum determined as The term containing the factor (-2i)® is the sum of all the determinants having two columns with that factor, Le., the sum of the determinants of the form where (r,s) is an arbitrary pair of integers taken from the sequence 1,2,... with r <5,86 Integral Equations Similarly, the term containing (Mi)? is the sum of the determinants of the form Ke Kee Kel = (A? Kp Key Kel Kip Kis Ku where rs, ¢ are the arbitrary integers taken from the sequence |,2,3,...n With r 0 and n>, then each of the terms of sum (7) reduces to some single, double or triple integral ete. Thus, we have b w pope lKEb) KE.E2)| af ean fe RG KOA a Da)=1 KG.) Kb) KG&)| K(Ep.E1) K(Eq.82) K(Ep.8g)/ db db byte (8) IK(Gs.81) K(G3.82) K(Es.83) eyes Where D(A) is called the Fredholm’s determinant. Similarly, the power series analogous to the series (8) may be written as Dayats Sep? x 2) ites (9) KES) Kibo) KE, Em) &, bp. +) K(G.8) K(G2.82) K(Gp.Em)) S1, 82 Sm where K{ KE mE) Km) K(Em Em) This is called Fredholm's first series,