Sudha’s Student Friendly Series.... SE Exclusive Special Feature BEATING THE MEMORY Formulae, Properties & Results to be remembered from all the chapters at a glance 510.016CHA INLLUN 1012159:lolalsd 93 ENGINEERING MATHEMATICS - lil for the THIRD SEMESTER B.E. COURSE OF V.T.U Dr. K.S. CHANDRASHEKAR M‘SC.,Ph.D Professor and Head of the Department of Mathematics National Institute of Engineering MYSORE - 570 008 In our endeavour to protect, you against counterfeit fake books we have put a mand the unique 3} pedal effeck gover on Some‘of our fast moving titles. The mand the w Tigpfays.a unique muflti golour effect from different when tilted un fake/hologram and wrapper does wt aS * Phone : 22216177 SUDHA PUBLICATIONS No.1, Annadanappa Lane, Avenue Road Cross, Bangalore - 560 002ENGINEERING MATHEMATICS - HT (FOR III SEMESTER) By Dr. K.S. Chandrashekar, MSe. /Ph.D. Published by M/s Sudha Publications, No.1, Annadanappa Lane, Avenue Road Cross, Bangalore - 560002 ZG. ES Edition : 2011-12 Copy Right : Dr. KS.C. & Sudha Publications Rights Reserved : This book or part of the book cannot be reproduced in any form including xeroxing. Rs. 330.00 Typeset by : ALLKIND Flat No.204, Nandish Park Apartment, Muthyal Nagar, Bangalore-54 Printed at: SATHYANANDA PRINTERS # 12, 4th Cross, Pipeline Road, Cholurpalya, Bangalore - 560 023.PREFACE Itwas in the year 1996, my first text book, Engineering Mathematics-III for the third semester B.E course of Mysore and Kuyempu universities was published. Since then the book has undergone three revisions as per the VTU syllabus which is revised once in four years. 1am completing fifteen years of career in the field of text book writing, without looking back. My esteemed readers on one side and the trusted publishers on the otherside are responsible for this memorable milestone. | am ever grateful to them in making, this possible. The current book ENGINEERING MATHEMATICS - II] prepared with all care is in accordance with the latest VTU syllabus. It also caters to the need of autonomous: institutions in Karnataka and other technological universities in the country [hope the book will live upto the expectations of the community of my beloved readers. My sincere thanks to two of my senior colleagues in the department, Dr, D. Mamta and Smt. G.V. Pankaja for their scrupulous scrutiny of the content in the book. I heartily thank Mr. K.V. Balakrishna of M/s. Sudha Publications for all the encouragement, publishing my books for all these years. I appreciate the highly professional computer type setting work by Sri. S. Raghunandhan and team of ‘Allkind’ and thank them. Talso thank the printers for the quality printing. I profusely thank the readers community, specially the students for the appreciative nice words through telephone and SMS I humbly invite comments and suggestions from all corners for further improvement of the book. August Ist 2011 K.S.Chandrashekar (Shravana Paadya) Mysore - 8 REWARD VIU students of the current scheme w.e.f 2010-11, scoring 125/125 in all four papers of Engineering Mathematics | to 1V Semesters (10 MAT 11, 21, 31, 41) will be rewarded with a cash prize of Rs.7,500/- by the author. Please write to the author directly along with attested xerox copies of marks cards of all the four semesters. : Achiever: Mr. Bharath M.V, a student from 2006 batch of E&C branch from PESIT, Bangalore, received cash prize during, 2008.SYLLABUS ENGINEERING MATHEMATICS - HII Code: 10 MAT 31 1A Marks : 25 Hrs / week : 04 Exam Hrs : 03 Total Hrs : 52 Exam Marks : 100 PART-A Unit - 1: FOURIER SERIES Convergence and divergence of infinite series of positive terms, definition and illustrative examples, Periodic functions, Dirichlet’s conditions, Fourier series of periodic functions of period 2 and arbitrary period, half range Fourier series Complex form of Fourier Series. Practical harmonic analysis. [7 hours] Unit - I: FOURIER TRANSFORMS Infinite Fourier transform, Fourier sine and cosine transforms, properties, Inverse transforms, {6 hours] Unit - 1: APPLICATIONS OF P.D.E Various possible soutions of one dimensional wave and heat equations, two dimensional Laplace's equation by the method of separation of variables, Solution of quations with specified boundary conditions. D’Alembert’s solution of one | wave equation. {6 hours} all these dimensi Unit -!1V: CURVE FITTING AND OPTIMIZATION Curve fitting by the method of least squares - Fitting of curves of the form y = ax +b, y=a +bxto, y= ac, y= ax Optimization : Linear programming, mathematical formulation of linear programming problem (LPP), Graphical method and simplex method. [7 hours]PART -B UNIT - V: NUMERICAL METHODS - i Numerical solution of algebraic and transcendental equations : Regula falsi method, Newton - Raphson method, Iterative methods of solution of a system of equations: Gauss - Seidel and Relaxation methods. Largest eigen value and the corresponding eigen vector by Rayleigh’s power method, [6 hours] UNIT - VI: NUMERICAL METHODS - 2 Finite differences : Forward and backward differences, Newton’s forward and backward interpolation formulae. Divided differences - Newton’s divided difference formula, Lagrange’s interpolation formula and inverse interpolation formula. Numerical integration; Simpson’s one-third, three-eighth and Weddle’s rules (All formulae / rules without proof) [7 hours] UNIT - VII: NUMERICAL METHODS - 3 Numerical solutions of PDE - finite difference approximation to derivatives, Numerical solution of two dimensional Laplace's equation, one dimensional heat and wave equations. 17 hours] UNIT - VIII: DIFFERENCE EQUATIONS AND Z-TRANSFORMS Difference equations : Basic definition; 4 -transforms - definition, standard Zetransforms, damping rule, shifting rule, initial value and final value theorems. Inverse Z-transform, Application of Z-transforms to solve difference equations. [6 hours] Note : In the case of illustrative examples, questions are not to be set.CONTENTS PART-A Unit-I_ FOURIER SERIES 1.1 Introduction... 2... 12 | ‘Cohvergence and Divergence of infinite sa 1.3 Periodic function Pe 14 Trigonometric series and Euler's formulae . . . . 1.5 Fourier series of period 2m 1.6 Fourier series of even and odd functions 1.7 Fourier series of arbitrary period . . . L8 Half range Fourier series... 2... - eee an 1. B 1.9 Complex form of Fourier series 1.10 Practical harmonic analysis 1.11 Additional problems . . Unit- 11 FOURIER TRANSFORMS ..... 00.0000 00 eee eee [119-156] 2.1 Introduction... . . ast 119 2.2 Infinite Fourier transform C niplek Pounee tranehoem). and Inverse Fourier transform... . . vere sd 2.3. Properties of Fourier transform (Linearity, Change of scale, Shifting and Modulation] 217920 2.4 Fourier cosine and Fourier sine transforms Inverse Fourier cosine and sine transforms with associated properties . . . 122 Unit-III_ APPLICATIONS OF PARTIAL DIFFERENTIAL EQUATIONS [157-194] 31 Introduction «= 5c s)ssemsoaead edi een ess Aare re mpc 157 32 Various possible solutions of standard p.d.es by the method of separation of variables. any pally 24 os #157 \Vantous possible'sclution of the one dimensional wave equation . «. .... 158 11s possible solutions of the one dimensional heat equation . .. . . . 160 Various possible solutions of the two dimensional Laplace's equation . . . 161 D’Alembert’s solution of the one dimensional wave equation . » 189 UNIT-IV : CURVE FITTING AND OPTIMIZATION... +++ [195-200] 4.1 Introduction... cee 2 195 42 Curve fitting by least squares method... . » 195 4.21 Fitting ofa straight line y = a+b 58 a wee cee ys 195 4.22 Fitting ofa second degree parabola y = ax? +bxtc . errr 196 4.23. Fitting of a curve of the form y = ae a3 eRe 1974.24 Fitting ofa curve of the form y = ax"... JT, RFs 088) 43 Optimization -215 4.31 Mathematical formulation of a ‘Linear Pcoeramming’ Problem (PP) .. 215 432 Graphical method of solving an LPP Pee Be eo a Senos Ue 22 2 2G 4.33 Preamble for the simplex method. as 3 é tot 233 434 Simplex method algorithm Fivga paranes s+ 290 PART-B UNIT-V NUMERICAL METHODS-1.........-.--.-++-- [251-296] GL. Introduction... on ee a ee ea EEE AES RODE y 6 aL 5.2 Numerical solution of algebraic and transcendental equations aks 251 5.21 Regula falsi method or method of false position .. 6... 6.22 ee + Ba 5.22 Newton - Raphson method 261 5.3 Iterative methods of solution of a system of algebraic equations nw 274 5.31 Gauss - Seidel iterative method 274 5.32 Relaxation method ........- : aan . 281 54 Rayleigh’s power method Lo ee Se eG . 289 Unit- VI NUMERICAL METHODS -Il... . . [297-366] 6.1 Introduction. ..... oe emo, Sei ee .297 6.2 Finite differences . whe eaneees one OOF 6.21 Forward differences ........- us ey i ear BI: 6.22 Backward differences 208% eee ee ee we Oe 63 Interpolation ©... 2.0.6.5 2 Suey 305 64. Interpolation formulaeforequalintervals. ..... ... eee ee 305 641 Newton’s forward interpolation formula ce . 306 642 Newton’s backward interpolation formula 6.5 — Interpolation formulae for unequal intervals 651 Divided differences we xe wm RR 6.52 Newton's divided difference formila 6.53 Lagrange’s formula for interpolation and inverse interpolation g 6.6 Numerical integration 23 , 6.61 Simpson’sone thirdrule.... 2.0.0... 6.62 Simpson’s three eighth rule 6.63, Weddle’s rule i eee eer Ter(T-VIT NUMERICAL METHODS -3 ...- 6-02 [367-414] 7A Introduetion . . ; rrr re 367 72 Classification of PDEs of second | order 367 7.3 Finite difference approximation to ordinary and partial derivatives . 368 74 Numerical solution of a PDE eo aa fo ele - 369 7.41 Numerical solution of the one dimens onal wave equation -370 7.42 Numerical solution of the one dimensional heat equation . 387 7.43 Numerical solution of the Laplace’s equation in two dimensions. . 398 Unit - VILL DIFFERENCE PQUATIORS RZ TRANSTORMS eee es 1415-472] 8.1 Introduction 2 : 2415 82 Difference equations mak elas 415 8.3 Z-transforms 416 831 Introduction to Z-transforms . 416 Definition and property - 416 7 ianstonmakeomestandard FUnCHGRS «cents ooalcin lest 417 8.4 Linearity property a9 8.5 Damping rule (property) 419 8.51 Some applications of damping rule 420 8.6 — Shifting rule (property) 421 (Right and Left shifting rules) 87 — Initial value theorem . 437 88 — Final value theorem 438 89 — Inverse Z-transforms 8 8.10 Solution of difference equations us using “ transforms waned BEATING THE MEMORY ALPHABETICAL INDEX - 473Unit -I Fourier Series [1.1] Introduction ble function ential Calculus we are familiar with the expansion of a different esof f(x) about x = @ isaninfnite In Diff #(-¢) inthe form ofa power series. Taylor's seri series in ascending powers of (2a) and Maclaurin’s series is an infinite series in ascending, powers of ¥. In many engineering problems it becomes necessary to expand agiven function f(x) ina series containing cosine and sine terms which belongs to alas of functions called periodic functions. In this unit we discuss various aspects of such series referred to as Fourier series. As a preamble we briefly present two concepts connected with infinite sereies of positive terms. ive terms 1.2| Convergence and Divergence of infinite series of pos If u, is a function of 7 defined for all integral values of 1, an expression of the form Wy 1 4g tou, +--+ containing infinite number of terms is called an infinite series usually denoted by ¥) or simply Yu, u, is called the n'" term or the general term of the infinite series. The sum of the first nt ns of the series is denoted by s,, That is, $= Hy ty tu Aseries Yu, is said to be convergent if lim s, = 1, where / isa finite quantity and D 4, is said to be divergent if lim s, = + . noe2 FOURIER SERIES Illustrative Examples Example -1 Let us consider the geometric series : a +ar+ar?+- Cd) and D iprot 2 Q Now if | r > 0 as n > © and from (1) we have, lim s,, = 72 (1-0) = 7"), which isa finite quanitity. i She Hence we conclude that the geometric series is convergent for | r | < 1. Next if r > 1, we have from (2), a(r—1) _ L lim s, = lim. e, since 1 —> %© when r > 1 noe na% Hence we conclude that the geometric series is divergent forr > 1. Example - 2 Let us consider the series: | +2+34++> +u4 (etd 8, = 1b24St rua Las fei) li = im PV. ims, = lim 7 Re Hence we conclude that the series is divergent. Example -3 Let us consider the se L gto y J vet us cons e series : us consider the serie aa +aat he nll termu, = | * n n(ntl) Further, 1 : 11 by partial fracti et, m, « = S- artial frac = Saud) Toad PY Patil fractions Now 5, = My tig tig te +,FOURIER SERIES OF PERIOD 2x 3 or 6 = 1- . 1 lim s, = lim [ae] 1-0 = 1, a finite qty. n> 6 ne Hence we conclude that the series is convergent. [13] Periodic functions A real valued function f(x) is said to be periodic of period Tif f(x+T) = f(x), T>0. k (constant), sin x, cos x are periodic functions of period 2m as we know from trigonometry that sin(x+2n) = sinx, cos(x+2m) = cosx Also if f(x) = k then f(x +20) Further we also have a property stating that A linear combination of periodic functions having period T is also periodic of period T. 14| Trigonometric series and Euler’s formulae The functions k, cosnx, sinax (4 = 1, 2, 3, ...) are all periodic functions of period 2z. Taking the constant k = aj/2, the linear combination of all the periodic functions is of the form bY a, cosmx + Tb, sin nx wet wet } are all constants is called a Trigonometric series. where 4,0, Hence any function f(x) expressible as trigonometric series of the above form must also be periodic with period 21. We shall assume that (2°) is defined in an interval of length 2x, say (¢, ¢+2m) and be considered as periodic with period 2a. Then we have, o+2n e+2n c+2n 1 1 a = fox) d. c f(x) si dx ay = 5 J fojdy a, 1 J FG) sinneds c ¢ e f Fox)cosnxdy 5 formulae. These can be The expressions for finding a, @,, 6, are called Euler established with the help of the following basic formulae.4 FOURIER SERIES con c42n 1 J cosmedy= 02 J sinnvas where 11 is a positive integer c+ 2n 2 f cosmecosmede= 0 J sinme sinnedx where mand rare positive integers, # 1 c+2n J sin mx cos medy = 0 where 1 and 1 are positive integers. e+2n e+e af costmedy = m= f sine da where 11 is a positive integer. Proof of Euler's formulae ayo 0 A+ Ya, cosa + Yb, sin nx - (1) Set nel Let f(x) = Integrating (1) war.tx trome to + c42n com, ‘ os [fords J Baee va, Joosmare 50, J sinmcty ‘ 2 wea” | wet” Mw on = 17 +040, by using (1) 4am we | fex) dx = ec} say cram Zs ‘, z aa J feo ax . +) Next, taking the expanded form of (1), mulbplying by cos mx and integrating wortyfrom ¢ to ¢+2x we have,FOURIER SERIES OF PERIOD 22 5 c+ on 42K c+2n % [foxy cosnrdy = 2 J cosmedraay F cosa cos xdx c+2n é tay J cosny cos 2vdx+-+ +a, [cos?nvdxt con e428 +b, J sinxcosmvdy+ty | sin2v cosmvdy+- Using, the basic results appropriately onto the R.HS of the above we have, 42a J f(x) cosnxdx = 0+0+4+- +4, "4040+ =a, 0 crm 1 a, = | £08) coma Q) Similarly, multiplying theexpanded form of) by. sin 1x and proceeding on the same lines we obtain c+ on b= fF £08) sim dy 4) Thus we have established Euler's forniutlae. Remark ; The constart term iit the series (1) is taken as a2. so as to make the formula derived for a,, valid for the particular case n = 0 as ell as for any positive integer 1 1.5) Fourier Series of period 2x Suppose we form the trigonometric series from f(x) defined in (¢, ¢+2n) with the help of Euler's formulae we cannot conclude that the series will converge to F(x). We can only say that when f(x) isof the form (1) the coefficients of the terms in the series are given by the formulae (2), (3), (4). We now proceed to state the conditions known as Dirichlet’s conditions under which theexpansionof f(x) asatrigonometricserieswillconvergeto f(x) atevery point ofcontinuity. 1. F(x) is single valued and finite in the interval (c, ¢+ 2x) f(x) is periodic with period 26 FOURIER SERIES 3. fox) has only afinite number of discontinuities in (ce, ¢+ 2m) 4. f(x) hasatthe mosta finite number of maximaand minimain (¢, ¢+ 2m). ) and satisfies Dirichlet’s Fourier series of f(X) in (4) respectively are called ‘Thus we can say that, if f(x) is defined in ( conditions, then the trigonometric series (1) is ¢ alled a ven by (2), 3) (6, €42n). The constants aj, @,,, Fourier coefficients. Remark : If f(x) is discontinuous at x, then the series (1) converges to t - 5 LF EE CE) | where FO). fC) are respect oly right hand and left haitd limits of f(x) given by Ie fOx4hy, FOC +0 hoo f(x-hi, hoo However at the end points f(x) converges to : fecy+fle+2m) | Note: Bernoulli’s generalized rule of integration by parts n integration ile as given While finding the Fourier coefficients, ir nrast of the problents we hnrae to perf ofa product cite te first function as a polynomial in x. tn such eases Bernowl’ below will he highly helpful. fuvdy=u fodv—w ffvdy desu” [ffods dy de Here are a few illustrative examples (ae) (@ ae a wor le é) 1 J xe dv =x |-D [5 | 2 J (x ta?) cosnrdy >, { sinnv ) (= cos nx sin nx =crtty (PME crea [OD |+2| a J fa h The following integrals will be useful in problems aw 5 (a cos bx + b siti bx ) a fel cos brdx = ei 2 @ sinbudx =f, Carsin by beos br)FOURIER SERIES OF PERIOD 2 x 7 The following values will have frequent reference in problems when n is an integer. 1 sina = 0 2. cosnm = (-1)" In particular cos (2n)m = +1, cos (2 +L) = ~ That is to say that cosnm = +1 when a is even and is equal to Livhen nis odd. Remark = {f is also possible to deduce a particular series from: the Fourier series of a given F(X) ia given interval. We have to substitute a suitable value for x in the given interval. Normaily we take the values to be either of the end points or the nriddle point. The resulting series will be equal to the value as discussed in convergence. WORKED PROBLEMS m=Nx 1. Obtain the Fourier series of f(x) =~ i Osx <2m. Hence deduce thal ae x gta pt aa f= FFE a, comme Yb, sinnx i) 124 nel me i 1 where y= [ £09 axa, = z J fx) cosmeds, by = = [ FO2) sina 3 é 4 n pr lpn 1 vd Von 8 acd Now, a) = ix = L| ay - = Qn —2n?)-0 = 0 Ww, dy = | dx aes 2 | oq CRW) 4 l 4 =8 x 5 cos dx. Applying Bernoulli's rule on (sin ny | -cos ny) | Oy Fag | BD 1) > |] n | ro L 0 = 2m va ie =< 3 fos", since sin 2x = 0 = Sind mit = ——5 | cos2m-cos0 |= 0, since cos 2um = 1 = cos0 2nn> | t8 FOURIER SERIES b, sind, Again by Bernoulli's rule, ‘ en —cos mx ) —sin ny tee hf tee (28-0 (=88 n> On ( ) 7 | ( y] e L \ Ah -1 on [ (n= 8) cos nx +0 2m 0 -1 -1 1 = {-meos2nn—mcos0) = (-4-4) = 2nn ) * rn " b, = Vn Thus by substituting the values of uy, 4,,. 8, in (i), the Fourier series is given by i n-x . sin nx f(x) = > To deduce the required series we put x = in the Fouvier series of f(x ) [ Note that at 0 or In, x = RLS of the Fourier series becomes zero and hence wetry x = 1/2 (0, 20)] sing | sin(3m/2) 2 4 sin( 5n/2) 3 + 3 tsindne 2 os for the function x° in —m < x < me and hence deduce that @ =p (i _ n : ii =c6 7 >> ‘The Fourier series of f(x) having period 27 is given byFOURIER SERIES OF PERIOD 2 n 9 4 7 f(xsy- y+ Ea, cos ity + Yb, sin nx -) wei nel where the Fourier coefficients a), a,b, ate given by Euler's formulae. 2 fxax x 3 ie om ae 8 i cos 1x d: oP cos neds =n {- ee pesinins) =(2x)| fre 2} a”) | ( ) . - [xcosnv]", since sin nn = 0 moe {m cos mm—(~m) cosim', since cos (~ im) = cos nm mor 2 = 3° 2K cos im au 4a(-1)" an 2 x x 1 r ot 2 - =) fice) sinmedx = 7 fe sin neds * =n sin nx { —cos nx ” art + ] wey 1 2 ~ (9 cos nm— 7" cos ne) +0+ % (cvisrancostin§ | ”10 ‘Thus by substituting the values of tg, Mr By ING) the Fourier series is given by f(x) Deductions : FO) 2 = ‘0 4(-1) 3 cosmx ie na ola w Putting x = 0 in (2) we get 2° # aga yl -™ 4 5 4 + cos 0 per =47 1 )= 0 =0; cos0=1 2 since f(m) = Wand cos nx = 4 w FOURIER SERIES -.(2) Ol)FOURIER SERIES OF PERIOD 21 ah 3. If f(x) = x(2n-x) in OS x S 2m show that 2n? cose | cos2x | cos 3x | 3 2 2 >> The Fourier series of f(x) having period 2n is given by f, 3 0 = faz + La, cosne + yb, sin ny 2 @ n= n=l Qn Qn Now ay = : J f(x) dx = 2 J (ame=22) de a 0 ae —ifo2 4) 17.3 se an? a(/2 = 2/3 Lt a a, =~ J 0x) cosmedy = es f (2m 27) cos ne dx 0 8 ,-! ’ (9) (an 20) cosmt) (9) =sinnx " wv 4 wm jh — 2 cos 2nt—2n cos 0) a= (cos 2m = 1 = cos0) on on 1 2 b= . J f(x) sinned = - J (amx-27) sin ne dy a 012 FOURIER SERIES yee ven (sy ) wo )h (ons) =2802 | (a¢-20f in n kon pn ws [oe + & cos nx | n I dy [(2Qar-2") iszeroat x=0, 2x and sin 2nm = 0 = sin0] 2 cos 2im—cos0) =~; (1-1) =0 mur" f(x) = naa Wt Expanding RHS by putting m= 1, 2, 3, . 2n® cosx , cos 2x soy =F -4(% i“ 2) To deduce the required series we shall first put_ x = 0 in 2). Since f(x) = 2nv-2? in 0 > The Fourier series of (() having, period 2z is given by a = . 0 3 . - oo f= 5 + La, cosnx + ¥ b,, sinnx vos(I) ae | ai :% i i = 5 J foxy dy, a, = ss ffC2) cosnedy, b= : fey Shcde oa =n “a Now, a)=— fe dr =a! on fer] 7 ag | 1 [ene | n| -a an Le MoT fener) % _ sinh an =t/ooe | o 2. 2" al 2 |} 2 me nr ag Ff or ® cos nx de = 1m We have the standard formula, wn Je cos bxdx = 5 (acos bx + bsin bx) week r a et 5m (-acos mx +irsin nx) | TM) (ay en La a -a Se = © cosny |, since sine = 0 = sin(—) x nia +i)14 FOURIER SERIES a : \ Oy = ——g og. [08 cosnm—e'™ - cos (nm) | (arti) -acosmm | a(-1)" an na = Ce Pe) t Ce ag ary m(a tn’) n(a +n’) _ 2a (-1)" sink ax n en?) 1% b, = 1 fe sinneds. on We have the standard formula, ya 5 é Je sin bx de = 5 (asin bx ~b eos bx) wth ; ” i max | b, i (-asinny—ncos nx) | = > sin \ "OR (aha ty ; ft =n ‘ = > > e ae cos NX I nan) | x The function to be evaluated between the limits —1 to m is same asini,, _ 2n(=1)" sinh an n(ar+i) Substituting the values of ay, 4%, , (1) the Fourier series is given by f(xy = inva | 2a tt sink ME ee s 2n ( 1" snk Om on ny an n(n) vot Re +I) —ax _ sinh = 2d (-1)" = 2Qan(-1)" Thus oe = Simla | 4 5 2a W cosmy + a1 sinny an 1) 4H 1 atu To deduce the series we shall put a = 1.x = 0 in the Fourier series ; sn 894 9 sine | 2 | since cosd% 1, sind = 0 molt tee? - | Ly" | be ‘0 ee ge Bet yoy EY de eee sit © anFOURIER SERIES OF PERIOD 2 x 15 m f-1 1 L 1 \ ie, = 1+2 gtog-set pe sink 14 142 1437 144 ) _ (loa ia ) =1-142[ 5-9 tay - | 1 1.1 TRUS!» Fin fa ( 10° 17> } 5. Find a Fourier series in (—n, 1) to represent the following functions . 2 fa) f(x) =x-27 (b) f(x) = +27 Hence deduce that >> (a) Period of f(x) = m-(—m) = 2mand the Fourier series of period 2x is given by 4 - ) : fs + Yo a,cosmx + Yb, sinsx where fi , i 1 - 1 Pfc ae, a= 7 [f(oeosnedy, 4,4 f f(x) sin ax dx es ay = =n on 1 % fa a t -ilfe x) [* =w)\\ _ -2n*" n)l2 2) 37 3 Jf> 3 a/2 = -?/316 FOURIER SERIES x”) cos x dy sinny fe )-a- yan sinnx [ i ” . 1 a 1 \ = 51 (1- 2x) cos nx] = (1-20) cosnm—(1 +2) cos nm mut om -4 4 4, 5 cosnm = -— (-1)" a ° a(-1yett a = Ato m wr 1 4 b= Pan jaea beh Fora) sin neds x _ = cost) (ayy ( ~8), (9) fea " } \ aw J \ =n 1f-1 2 2 7 | (ATE) cos 1 (-2-2) cosnm— *, (cosnm cost) ™ " on I 2 by =~, (2rcosnm) =, (-1)" 2 b= =(-1y"3 ny (OU ‘The required Fourier series is given by yee wd Ei -1 A ate By yp cass + 2 sin nx () 1 ‘To deduce the series, first let us puty = 0 which is.9 point of the interval (-1, 1). Hence (1) becomes ® eryttl gy O=- 140FOURIER SERIES OF PERIOD 2 x 7 .() or Ba -ata-o Next let us putx = min (1). Since f(x) = x-x7 is defined in- m < x < mthe value of f(2) at x =7 being f(x) is given by Lf(-m)+f(m)] which being 5 [tay 4c8 c Hence (1) becomes ae: - (aaytt -P a +4 y 1 cosnn +0 {oon bs a+] 2 sori ie, 4 5 =47 Ct, since cos nm =(-1)" ~ 1 w eo-1 1 1 42 fee »-Q) = jtatate - 3) (b) Fourier series of f(x) = x+2° in —2> The Fourier series of f(x) having period 2m is given by % ° 2 oS i f(x) = 27 Y a, cosmx + Yb, sin ite het sie <7 [* on 1 Now, ay == fea dem | faa de + { to) ae 4 D ® 4 m on | je, ay = Pade [anv a) 0 ca 7 Qn -112) .Jox-% | - + | nx - | 0 7 La 2 _ og? 2 \l_lie iy = — 0) + (4n2 = 20?) =| 2m [paar \ A y/2 = 72 2 % a a, 2 f Foxy cosmx dx = 2 | f f(x) cosine des [ f(x) cos med rt x 0 0 1 7 on a, = 24 J ccosme des f (2n-2) cosnedy 0 fa aa [=(82)- mt 1" aa Aafanil he 1 [comes (8) — (oy (=22822) L~ a -| rd h where we have applied Bernoulli's rule to each of the integrals.FOURIER SERIES OF PERIOD 2 0 19 1 34 1 4 : a= J [cosmx] -fcosix] |, since sini = 0 nm J ae = sind 1 = 1, (cosmm-1)-(1-cosnm), since cos = 1 = cos 2nn nie 1 =2 = 3 (-2+2cosmm) = —5 (1~cos nm) nr mut -2 a nt Je at a-(-1)"} 2n 7 an 1 b= 2 J Fexysinnxdr = | Dy es sinnxde+ J £2) ‘anal 1 n| | 0 oO n m on | b, = UI xsinnedy + [ (2n—x) sinnedx *la : n Me (one -a) fees =I Lo } Lon fl 0 . Onan [28H] = 4} n | / —t © . an = | tx cos nx J +04 (2n-x) cos | -o! nn | 0 IK he | b= (mcosnn—0)+(0-—m cosnm)| = 0 ’ b,=0 The Fourier series representation of f(x) is given by f= 5+ y = {a-(-1)"| cos nx _yy. Ji-(-1) = 2 ifn is odd But TICES = 1141) = 0 if mis even 8 f(x) = 2 - 2 cos nx20 FOURIER SERIES nm 4[cosx , cos3x | cos 5x re fey FEE EH To deduce the series let us put x = 0. Then f(x) = 0 since f(x)=xinOsxem Hence the Fourier series becomes 7. Obtain the Fourier series for the function 4 [em in -n> f(x) isdefined in (—m, ™) and the Fourier series of f(x) having period 2n is given by % = < faye Y a, cosnx + & bysinne Cl) n=l n= 1% 1 0 n | ay = SPs) ae = [fon ax+ J fox) ae 0FOURIER SERIES OF PERIOD 2x 241 a, . = 2 J £02) cos nade {o t | J F(x) cos nde + J f(x) cos nxdx} # a | i | v+ J xeosnxdx @ sin mx = cos nx : -l4 . " v | 0 0 where the integration is carried out by Bernoulli’s rule in the second integral 1 nl a . = 5 [eos mx |) = 5 (cosm—cos 0), since sin mm = 0 = sind ~ muy mm -1y | 4, = 1-(-1)") net "pn? fo n F(x) sinnxdx + f f(x) sinned r 0 " 35k deca | {i n J -nsinnxax + [ Aina | | [ssn 1 ny i = cos nx | y. csinax “57% =f | 0 ie 1 oO nt = Leos nx] -[xcosm| since sin nm = 0 = sin0 my T 0 1 (cos 0 - cos nm) ~ (cos nt 0) mn22 FOURIER SERIES my 1 b= 7 | 1-cosnm—cosnm| =~ (1-2 cos nm) nn n 1y a =hla-2(-1 b= rR" Substituting the values of a, @,,, 8, in (1), the Fourier series is given by es a : i fixy=— 4d - Ly (nay jcosm + 5 + (4-2(-1)"| sina ‘ 2 a i" net wet To deduce the required series let us put x = 0) in the Fourier series [tshould be observed from the given f(x) that x = 0 isa point of discontinuity and hence the series converges to J [roar |= 5 [ort x) a Because to the right of 0, in (0, ),f(x) = x and (0°) = 0. Also to the left of 0, in (-m 0),f(*)=—% and f(0") = — Hence the Fourier series becomes a |, since cos 0 = 1,sin0 (1-(+1) = 0 when ni is even Bat 1-(-1) = 94 _(-4) = 2 when nr is odd x 8 Hence we get 7 y 5 (2) or 3, Thus the sum of the reciprocal squares of the odd integers is W/8 8. Obtain tite Fourier series for the function 0 in-n > f(2) isdefined in [—n, 1] and the Fourier series of period 2n is given by a ; f(x) =F 4Y a, cosmx + 5 b, sin mx (a aA net 1° jo | wong Slaten s [pos ae J pes) a -1 oO 1 | oa » J snvas) =2 J sinzds ‘ L J <1 2 a we — cos x (cos n-cos0) = - 1 ( 1-1)= Con T 7 a/2 =n ( 4 ] 4, = 2 J £00) cosmede = 1) f 0» comedy + {Bey cos nx dx! oe Ls 0 x 1 ‘ [sin cos me dx. D Using sin A cos B = 4 sin(A+B)+sin(A-B) x 1 . a= i {sin (xt) +sin (a—mx)) de i a! 0 i 1 a J 'sinCltnjetsin(1—n)x! ax é k J isin rs 1) x-sinGr-1y x dx 2n o if oe s(n-1)x = 1] scos(nt dye | cos(n—1)x | diem ear 2 n+l n-1 |, 1f-1, 1 | == cos (+1) cos + | cos (1-1) 2 cos 0 2n [n+l n-1 | Using cos0 = 1,coskn = (~1)* and rearranging,24 FOURIER SERIES yet, ge] fais ‘wed (=1 (yy 73 [a walt (aro) “hn n-t] med 1) (1 Se Now we shall find a, when 1 = 1 Thatis to find a, 1 1 Wehave a, =— J siny cos nxdy _— 0 1 1 i 1 1 F sin 2x Putting = Lay = © Joins cox = = J dx Q 1 | -cos2x [P| -1 1 ie. "= on = s Le pe (eos A= cos) = 4 1-1) =0 a, = 0 and a, = SP fae ceay"|, for net rcs, 7 i iis 1X b, =| J fox) sinnedx = 4 | [0+ sin nx dx + J sinx sinned } : les x 1 fe, b= x sinnxdy 1 Using sin A sin B = ‘ cos (A-B)-cos(A+B),FOURIER SERIES OF PERIOD 27 25 ® fl bo Jy (cos (xm) cos (x tne )) dx 0 1" 5 J cos(1-nyx-cos(Ltnya dr é it — J cos (1) x= cos (nt 1) x dx, since cos(—@) = cos0 0 where 1 #1 _ 1 fsin(n=1)x — sinQrti1yx On n-1 n+l 0 b, = 0(n #1) since sin kx = 0 for integral values of k. Now we shall find b, when n = 1. Thatis to find by * 1 5 Wehave b, =~ [ sinx sinner dy # ™ 6 a a Putting = 1b, = z fi sine sinzdx => f sin? vax tn 0 » 1 1a 1 sin 2x ie, bb J 3 (1-008 2) dx = [* 2 [ 1 Itt = 5, (B- 9) = 9 since sin 2m = 0 = sind by = 1/2, b, = 0 for #1 Let us consider (1) in the form, % 2 +My cose + XY 4, cosax+by sinx+ Yb, sinnx n=2 n=? fa) = Thus by substituting, the values of the Fourier coefficients, the required Fourier series is given by zd \ a 14(-1)") cosnx + > sinx 1 Poet 2 nea) 2 Yo deduce the required series we shall first put x = 026 FOURIER SERIES -|i4(-1y")-1+0 1) #8 np mt Also (0) = 0 aswehave f(x) = 0 in-m 2 : i, o- -¥ sh (2) cos a n=2, 460-1 ie. Bee lL dearth ok este 3 \ i TS g On ag cos tgs emma hon But cosa = —1 = cos3m, cos 2n =FOURIER SERIES OF PERIOD 2x 27 9. Analiernating current after passing through a rectifier has the form [lo sinO for 0<0>. The Fourier series of period of 2m is given by ig : 1=f(O)=5 + Ea, cos 8 + x b,, sin nd 2) We have to find the Fourier coefficients by using Fuler’s formulae for the interval (0, 2m) Qn Qn an 1 7 1 1 a= { fceyde, 0, = : J fC) cos nda, b, = = f £c@) sin no ao 0 0 0 Each of the above integrals after splitting into two integrals in the range (0, 1), (1, 2n) and substituting for f(6) will give us i iy. Ty a= 2 J sinoao, 0, =? J sindcos nado, b, = 2 f sinOsin no do 0 0 0 These integrals are sameas in the previous problem and hence the Fourier coefficients are as follows. 2h 1 0 0 F = a, = 5 1+(-1)"| if a4 La, =0 ™ m(n-1) I b= 0 if mz tand b= 5 Thus by substituting these values in (1) the required Fourier series is given by S 1 | h +z 2 1+(-1)"| cosn® + 2 sin®28 FOURIER SERIES 10. Obtain the Fourier series for as of 1 [-k i (OY pence deduce that Z= 1-44 4b - FOO> Vek in (0, m) 4 a*5 so The Fourier series of f(x) defined in (~1, ™) hi 4 faye 2+ ZY ajcosme + Yb, sinnx (1) “ med ‘ a=, ffodde= 1S pa | ' [fen a) -n mle |‘ 7 ay = hy | ~kav + Yeas 0 = iy ma 0 a, = & =(047)+(m-0)! = ™ a/2 =0 17 |{ " | a, = zie cos eds = © Lee cos mxdx + [ f(x) cos nx ax | ™ o | 0 x a,=2 | J -keos nxde + J keos nx dx x d 0 © k sin 3 s IX is [| al x] tn : iF =0, since sin0 = sin mt a 0 n (0 } b= 1 J foo sinnyde = «! J fo sin pede i f(x) )snmvat 0 ™ | b= 4 il. —ksin nx dy + J kin ax |FOURIER SERIES OF PERIOD 20 29 ° = k | cos nx cosmx | | b= SOs. qh ea ese] [se] | Le 0 by k =~ |(1-cosnm)-(cosnm—1) = — (2-2 cosnm) x mn Pk To fa5e b= = 1-C1y"| Thus the required Fourier series is given by f(x) = > 2 j4-(-1)"! sinny aw. [l=C41) = 0 if mis even But THC" 91 -(-1) = 2 ifm is odd F(x) sin nx = 4k [sin | sin2v | sin Sy | sin ay Fe ia 2 3 4 To deduce the series let us put x = 7/2 Then f(x) =k since f(x) =k inO> The Fourier series of f(x) defined in (—1, =) having period 2n is given by » ; . ‘0 “ > Ree yt YE a,cosmx + Yb, sin nx f(x) =30 FOURIER SERIES “fh a (5 im | % = Jroyar= 1 f pcx dv + J f(a) dx, ~~ -% [ox 0 J 6 if J aga k lf [1+ \are [[1- =) ar 9 | T és ™ ex ay = 2 \0-(-m4)+ (Hon) =O) = 0 a/2 = 0 it a, =F £C2) cos meds a i 7 ) abl | p(x) cosmedy + J £0) cose dr) "La a ) 1 [Ge . 2) * 2 | = Slavs = cosmvdx + f [1 - cos me dx x t : } an } | { oO 1L 2x) (sinnx) (2) { -cos 3{l-28-OC8) ( ) lel a ( = {(—3) (Sr) 13) Soar | 7 n n a” yp) But sinnm = 0 = sin0 2 1 a = 2 1 -cosnm)—(eosnm -1)} = Ag (2-2eosim) renFOURIER SERIES OF PERIOD 2x 31 n {‘ 7 | b=} f pay sinurdy = 1) fpcry sinmedy + J pcx) sin near « n + Le 0 | 0 1 if + = sinnx de + f (: = 2) sinus - D ") | Lier | | (2) CF HES) “U1 4 2) (eos 0] f, 2) = U + a | (osm) + [fa | cos | i 0 —m -1 = |1+cos mm -cos nm—-1| = 0 mH #0 nt Thus the required Fourier series is given by oe 4 f= Yay IH 1)"| cosmx n-1 TH if it iseven ifn isodd wert es Ji-G= But = 1-(-1)" = feces cos nx cosx | cos3x cos 5x + + + 8 f(x -4/ or (x) PoP 2 232 FOURIER SERIES Fourier series of even and odd functions Definition : A function f(2) is said to be even if f(—x) = f(x) and odd if f(-x) =-f(x) 2 6 ; For example,x2, x4, x°, ...cosx are even functions and x, 2°, 2°, ...sinx are odd functions. Property-1 : ‘The product of two even functions and that of two odd functions is always even whereas the product of an even and an odd function is always odd. 2 Jo Cx) dx if 6 (x) is even 0 +a Property-2 : J 6 (x) dy = eg 0 if @ (x) is odd Now, suppose the periodic function f(x) és defined in the interval (~ 1m, ™) then the Fourier coefficients are given by 1 t x 1 1 1 a a % = 5 J fx) de, a, = : J F0x) cos meds, by = 2 Jf 02) sin nd on -n 7 Let us examine the following two cases Case-i: f(x) isaneven function F(x) cos nx being the products of two even functions is also even and f(x) sinnx being the product of an even and odd function is odd [Property-1]. Now applying property-2 to these integrals we have x 1 2 [foo de, a= 0 J f(x) cosnedy,b, = 0 0 0 My = fC) isan odd function nx will bean odd function and f(x) sinmx willbe an even function. Now 2. we have a =0, @,=0, b, = 2 J Fx) sin nx de d Thus we can conclude that when x € (—7, %), if f(x) is even b, = 0 and if f(x) isodd a, = 0,4, =0FOURIER SERIES OF EVEN & ODD FUNCTIONS 33 [o(x) in -n> ‘The Fourier series of f(x) having period 2n is given by % = ) fox) = 3 +a, cos nx + Yb, sin nx wer) nel hall check f(.°) = sin( mx) for even or odd nature. f(=x) = sin(- my) = —sinmx = -f(x) f(-x) = -f(%) and hence f(x) is odd. Consequently a = 0, 4, = 036 FOURIER SERIES * & 2 a 2 b= = J f02) sinnedx = 7 J sin mx sin awd 6 6 in A sinB = ; cos (A-B)-cos(A+B)| we get Using | cos (m—n)xX-cos(m+n)x| dx m-n men = sin (m+n) x lb Note: [fis important to observe that sin k 1 = 0 only when kis an integer. Since m is not an integer by data, sin (m—i1) x, sin (m+) are not equal tozeroand the sinplification is carried out as follows L ‘, 4 1 . b —— |sin(m-n)m-sin0} — —— sin(in+n)n— in 0} % man met 1 ’ 1 i J — sin(mm-itm) — sin (mm +i) na |m-a mtn 1 —— (sin mn cos im —cos nm sin nm) L ~—— (sin nm cos nn-+cos mm sin nm yt m+n J 1 J a2) sininm cos nm | —— - —-— || ] m-n m+n i (Here sinnm = 0, since n= 1, 2, 2) 1{ 2n_)] b = — } sinmmn cosim | ——— |\ non m—r? | p = 2nd sins 7 nt — at Thus by substituting the values of 4), 4,, 8, in (1), the Fourier series is given by = 2n(-1)" si ; sintnx) = Sees Zl SIN ET cin nxFOURIER SERIES OF EVEN & ODD FUNCTIONS 37 13, Obtain the Fourier series in (—m, ™) for f(x) = xeosx >>. The Fourier series of f(x) having period 2n is given by 4 “ : 0 FO) = a+ YE 4,cosme+ Fb, sin nx .() et net We shall check f(x) = xcosx for even or odd nature. f(-x) = —x cos(—x) = —xXcosx f(x) = -FC) and hence f(x) is odd, Consequently ay = 0, 4, = 0 2 ‘ 2 , J £0) sinned = J xcosx sin nx dx 1 fn b, = 0 0 at ie, b, = 2 J vsin nv cos xd @) 6 Using sin A cos B = , |sin(A+B)4sin(A~B)) wehave, 27 4 Lt Js “5 isin(axtx)+sin(nx—x) dx 1 |" , =o fox sincn+i)xdy + [x sin(n-1) xdv} i l Applying Bernoulli’s rule to each of the integrals, cos(n A) sin(n+1)x) [ jx- = | n+l (n41y h _cos(m=1)x 4 _ sin(n=1)x n-1 (n-1 ir [wen | 1f{-1 - 1 = [peri meosterd) 0 a [meos(n—1) m=O] Here sin(+1)m = 0=sin(n-1)m, since n= 1, 2, 3, ...38 FOURIER, SERIES feas(n+i)m - cos(n-1)m | b= +l n-1 | feo, er '| utd | fa \ ( -(-1y" 4 a CM Viet Pat | “ay ea 1y (1 #1) wat We shall now find &, when i = 1. Thatisto find). Let us consider 6, as given by (2) 1 a f . dee 8, = 7 J xsinaxcosx dx. Putting = 1 we get, 3 * 2 " 2 . sin 2x b, =< J xsinxcosxdx =~ f x dy T 2 3 0 a J wsin avd. i LT = Gq Ly cos2vy) -» sin2m = 0 = sino “1 =< (m cos 2n-0) cos 2n = 1 Qn b= -1/72 We shall write (1) in the form 0 fixy=5 4 a,cosay+b, sinx+ Sb, sin nx ‘Thus by substituting the values for the Fourier coefficients we have, -1 x cosFOURIER SERIES OF EVEN & ODD FUNCTIONS 39 14. Expand the function f(x) = x sin. asa Fourier serie Dedduce that + Y a,cosmx+ Yb, sinny qd) n=l We shall check f(a) = xsinx for even or odd nature. f(-x) = (-x)sin(-x) = (-x)(-sinx) = xsinx F(x) = f(x) andhence f(x) iseven. Consequently b, = 0 8 & Pe: J fx) de pa, = 2 J fOr) cos nx dx a = i) at ay = 2 J xsin da. 0 2 ” a = x(-cosx}—(1)(-sinx) |. But sina = 0 = sind og a ie, ag = (reosn-0) = 2 8 ay/2 = x a= 2 f xsinx cos nx dx Q) " 0 ty. Using sin A vos B = 5 |sin(A+B)+sin(A-B), we have, 1 1 4, = x+y sin(xtax)+sin(x~ nx) dv 0 . a No 1” 1 =— fix: {sin(l+m)x+sin(1-a)x| dx. wed But sin(1—)x = sin ~(1—1)x) = ~sin(n—-1)xFOURIER SERIES a= * . 1 ak J xsincnetyady 5 J xsinn 1) xs 0 0 to each of the integrals, _ ay. ines qnt1y 0 Lyx n (res D Gy asin DP iy ety | nt aay iy Applying Bernoulli's rule =sin(n-1)m; sind =0 x cos (n+ 1) x meet ye] a-1 0 H Ty 2] xeostu=Da yr nT 1 [_m_ egs¢n=1) 2-0} 1 , JT (recone 1-01 = ran cos(Ht1)m | cos(u—1)m ~ n+ n-1 qt yi 2-1 EE, CD | since cosk m= n+l n-1 (ay? ~ ntl But (iy! a a where n #1 We shall now find @, when = 1. Thatis to find a Let us consider @,, as given by Q). Putting "= 1 we have © ™ fa sin 0 2 =? J xsinx cos xdx = dy tn sip 0FOURIER SERIES OF EVEN & ODD FUNCTIONS 4 : 1 a xsin 2x dx = — 2 (As iit 6, of Problent - 13) ye Substituting b, = 0 in (1) wehave f(x) = > + Ya, cos nx nit - ay = ie, POX) = og) Faycosx + Ya, cosy Now substituting the values of a9/2,a,, 4, (at # 1) the required Fourier series is given by , 1 * 9¢-1y"*! xsinx =1- 5 cosx# Y > cos nx ? n=2 wl To deduce the series, let us put x = 2/2. x nd mom loom (-1)"" nm zg =l-yjcosst2 & c pein g=1- 9087 +2 Bay cos . wm r 1m ie, since sins = 1,c0sy = 0 ie. 1 cost ---| 2, a5 0 ie, , since cos % = — 1, cos 2m = 1,cos (3n/2) = 0 15. Sketch the graphofthefunction f(x) = |x| in —1 Sx Sw andobiain its Fourier series. ‘4 Hence deduce that 7 5 o> F(x) = |x| in -2 > f(a) isdefined in (—1, ™) and the Fourier series of period 2n is given by44 FOURIER SERIES fi . 0 . * FOX) = 4 + Y a,cosnxy + Fb, sinnx n= n=l We shall check (+) for even or odd nature by writing f(a) as follows. Interval ofx | (—n, -n/2) C 1/2, 0) (0, 2/2) (1, f(x) MAX | n/2 n/2 n- A a i ae m, 0) Be fry = jy ( } in (0, m™) where $ (x) = "+x or 2/2 and y (x) =a-x or 1/2 (-x) = m-¥% or m/2 = w (x) and hence we conclude that f(x) Consequently b,, = 0. * 1 2 2 ay == ffi, a= > | f(x) cosmrd " 0 m 0 ) (nr © | ali aes Fyeods) =? J Fav + [(n-xy ax a fe ee | a,/2 = 3n/8 a2 a,= oe 0 n/2 e [i F(x) cos nx dx + Thos coma 2 1 Fe ca | | 5 cos mx dx + J (m-x) coon wh v7) T) x is even.FOURIER SERIES OF EVEN & ODD FUNCTIONS sins | a= ay a[a n|\2 T mm 1 HB | \ =2 vn 222. (cosnmcos”™ mu ai nm 4, = (-1)" - cos * | or 2 aAIN Thus by substituting the values of ag, a,, 6, in (1), the required Fourier series is given by Even and odd nature of f(x) defined in (0, 2%) F(x) is said to be even if f(2m=x) = f(x) and odd iff(2n—x) = — FO) We note that cos x is an even function since cos (2m- x) = cos x and sin.xis an odd in x. function since sin (2m—x) = Further we have the standard integral property : a 2) flx)de if f(x) is even 0 da f f(x)dx = a 0 if f(x) is odd If f(x) is a periodic function of period 2n defined in (0, 2) then the Fourier coefficients are given by a on Qn [rae a, = J fxvcosmeds, b, =} f F(x) sin nx de 0 0 * 0 Let us examine the following two cases. Case -i : f(x) is aneven function. F(x) cos 2 xbeing the product of two even functions is also even and f(x ) sin i being the product of an even and an odd function is odd.46 FOURIER SERIES ‘The Fourier coefficients with the application of the integral property becomes 2 & == | Foxy adr, a x == | fix yeosnrd, b, = 0 a 0 a 0 Case-ii; f(x) is an odd function F(x) cosnx will be an odd function and f(x) sin x will be an even function. Accoidingly the Fourier coefficients with the application of the integral property becomes X J fC) sinny dy 0 Thus we can conclude that when x © (0, 2m), if (4) isevenb, = Oand if f(x) is odd a, = 0,4, = 0 Further it should be observed that the results in respect of the Fourier coefficients involving f(x) defined in (0, 2m) are same as in the case of f(x) defined in (—1%, 1) where f(x) is even or odd in the relevant interval. o (x) inD«x< Also it may be noted that if f(x) = it Goins 7 A we say that f(x) is even if @ (2n-x) = w(x) and f(x) is odd if o (2n-x) = - w(x). WORKED PROBLEMS s three of the already worked problems using the [As a matter of comparision we briefly d concept of even and odd firnetions} Referring to Problem-1 f(x) = ‘in (0, 2m) () Ma(m-x) mtx -(m-x) f(2n-x) = 2 =o 2 =-f(r) f(x) isoddin (0, 21) and hence a, , a,=0 © b, =3 f fOdsinnxde ° 1) 1 b= J (n-x)sinnxdy = 1 on integration, 0FOURIER SERIES OF EVEN & ODD FUNCTIONS 47 Referring to Problem-3: f(x*) =x(2n-x) inO< xs Qn f(2n-x) = (2Qn-x)(2n-2n-x ) = (Qn-x)x = f(x) F(x) isevenin(0, 2x) and hence b, = 0 2 ‘ z & ty == 1 firiae, a, == J F(x) cos nx dx 1 2 fe Pitty = an” i ay = | (2nx—3?)dx = “>, on integration 0 x 2 2 -4 a, = f (2nx- x") cos nx dx = —5, on integration. " % w 0 Referring to Problem-6 f(x) = { ae lot fx) = {9 x in 0 =n w(x) =2n-x in nex < Qn 0 (Qn-x) = 2n-x = W(X) f(x)isevenin (0, 2r) and hence b,, = 0. x © 2 ay= 2 fread, a= 2 [so eosneds 0 0 ® > ay ~~ J xdx =m, onintegration O x 2 0 “ a= f xecos mx dx = ~(-1)"-,0n integration. fe = } 17. Obtain the Fourier series representation in (0, 21) of the function f(x ) defined by 2 AGS ool © in (0, m) |-(an-x in (nm, 2m) >> Inthe given f(x) let @ (x) = 2 and w(x) = (2n-x Now (2n-x) = (2n- -w(x) f(x)isodd in(0, 2m)andhence a= 0, 4, = 048 FOURIER SERIES 8 rk : J f(x) sin nx dx = 2 j x? sin nx dx a a + 2[%3 | b= Wm _yyrdy 3 \¢-ayt-1} mn aa ‘Thus the Fourier series representation of f (x) is given by f= = [Poets 4 nal] sin nx TH 48. An aliernating current after passing through a full wave rectifier has the forn 1= 1, | sint |, 0 By data I= f(f) = fy | sint | in (0, 2n) _ [40 sint in (0, @) fey TSEC 1 sine im (ae, 2) since sin f is positive if 0 < t < m and negative if x < f < 2a (oC =A sing in (0, 7) bet P= fC = Fy (yy = 1 sint in (, 2) 9 (2n-t) = Iysin(2n—t) = —Igsint = w(t) f(£) isevenin(0, 2x) and b, u © 5 2 2 a= 5 [fond 4, =5 f Fee) cos ant at 0 x 0 21 4I ® popatypond 2% 24, 2 0 = 4 sin tdt = —2 [cost = 0 = J Jysin (dt = —* [cost JjFOURIER SERIES OF EVEN & ODD FUNCTIONS 49 ay / 2 = Uy 1 J sini cos nt dt 0 Refering to Problem -8 for the integration process we have “2g a, = n 3p ~ |tH(=1)") form #1 and a, =0 n(m-1) We have Fourier series in the form. 4% 0 (= 5 + Ea, cosm + Yb, sinwt not nat 2h ~2% f(th=—+ = j1+(-1)") cos at mys OP-1) 2h 4], = Thus f(t) = oO 0 yr sgeut TO ganae. WAT To deduce the series let us put # = 0 fO)+FO2m) _ 0+0 fog) = Het iten) eee. f(0) 2 2 0 The Fourier series becomes, 2h Al, - o=—- s A 19. Obtain the Fourier series expansion of the function [x mocxen x)= FOO om fn mc ye Om I Hence deduce that pti >> The Fourier series of period 2m is given by50 FOURIER SERIES + Y acosny + Yb, sinne fQ)= n=l w=] In the given f(x), let @ (x)= x and w(x) = x-2n 6 (Qu-x) = 2n-x = -(x-2n) = - W(x) f(x) is odd in (0, 2m) and hence a= 0,a,=0 ® ® 2 . 2 . b, == | f(x) sinnxdx = = x sinnx ax ® 0 oO of ( x = cos nx sin a -2[+[ ; J-eo[ zi }| Ww dh =8 a b, = = meosnm = 55(~-1) wet p = 201) # n ‘Thus the required Fourier series is given by 5 Brel Fay= y 7D sinnx To deduce the series let us put x= 1/2. Then f(x) = 17/2 since f(x) =x in (0, m). Hence the Fourier series becomes ° 1 % (-1)"' nn =2 > 27? en NG. Expanding and noting that sin ( 32/2) = -1, sin (50/2) = 1, we getFOURIER SERIES OF EVEN & ODD FUNCTIONS 51 2 mre = in O< x < 2m, show that cos nx 2 ~ f@i= 44 2 . Hence deduce that 122 T >> The Fourier series of period 2x is given by % = - Q . f= Fe ET aycosme + Yb, sin nx net ned Consider f(x) = ( =) fOn-x) =| Bre f(x) iseven in(0, 2n) andhence b, = 0 é ‘ 27. 2 © frends, a= 5 J fxjeosneay 0 0 p é 1 t 2 af 2 _ (a-x, | (may P 1 el 4 whl =9 |, 7 on w)= 6 ns Ww 3 ® i) aly 1 cosnxdy = — f (n-xPcosnedx Qn | oxy 3 o52 FOURIER SERIES Applying Bernoulli’s rule we get, 2 ( sinnx ) _. (=cos nx) (-sinnx ) ft (x-x? [ a |e ayia i ae —— Ih \ ) ” 1 On 1 “1 { = [(n-x)cosnx = 5 (0-") = nie 0 mr ne a, = Vv? Thus the required Fourier series is given by ~ 12 2 2 " cos 1x #2 7 on foxy (3 To deduce the series, we first put x = 0 +(-02y" _ 2 4 . id 2 ir 1 Thus i) 3 Next let us put x = ™ in the Fourier series. cm cos nm -0 (-1)" O=a+ Or at 12 1 r 12 rv mus Hebe S45 - G4 (ii)FOURIER SERIES OF ARBITRARY PERIOL 53 7| Fourier series of arbitrary period A function f( 2) need not always be defined in an interval of length 2m only. When, the length of the interval is other than 2r, we shall denote it by 2/. A general interval of length 2/ be (c, ¢+2/). [tis important to note that the sine and cosine functions f 1: of the form sin | a and cos ( =] are periodic functions of period 2. It is ae justified as follows. ( Let F(x) = sin Ur} G(x) Then F(x+2I)=§ 5 | F (rez) | = sin | 7 +2 | \ ) {me = sin | I J-Feo Similarly G(x +21) = G(x) As already discussed in the article 1.4 the trigonometric series is of the form mS __ (an) : a, { 1nx ) zt cal ; Jz b, sin’ | If f(x) dialined in (c, ¢+2/) satisfies Dirichlet’s conditions then % = Kim flys 7 + ¥ a,cos es \+ b, anh | is called as the Fourier series of arbitrary period 21. Proceeding on similar lines as in the article 1.4 we can establish Fuler’s formulae for the Fourier coefficients in the form c+2) J f(x) dvya, © e+2t b=) Pfc sin | x, nm \a c+at [Fo cos | Working procedure for problems D> Ifthe period of the given function is other than 2n_we first equate the period to 2 and obtain the value /. 2 We then write the appropriate Fourier series and compute a, 4, 6, associated with it.54 FOURIER SERIES > However if f(x) is defined in an interval of the form (~1, 1) or (0, 2) we cancompute dy, 2,,, b,, using the concept of even and odd functions faking the following table into consideration. | Nature & condition | My | ay b, of fe) 1 | 7 Even function |, | zt . fl-4) =f) J poy ae!% [ p02) cos Be 0 | or | Mo hy t | | festexy=fory | Lo FE | Odd function [2 ! | flex) = FO) | 0 | 0 7 LC sin de or i 0 f(l-x) = -f@) | WORKED PROBLEMS 21. Obtain the Fourier series of f(x) = |x| in (“1 1). 1,1,1 c Hence show that +54 = ie x 8 55. The period of f(x) = 1=(-1) = 21. and the Pourier series of period 21 is given by a 8 7 , f(x)= > + ¥, a, cos = + ¥ 4, sin oa ...) nel n=1 We shall check f(x) = | x | foreven or odd nature f(-x) = |-x] =| 2)=fC) Hence f(x) is even and consequently b,, = 0 ! ' 2 tz 2 ay = 2 [pry aes a, = 7 Lf edeos “Pax 0 0 ! f27 m=z hee=F| | =} (P-0j=1 0 “4 ay/2 = 1/2 a " 1 2 = 2 fx cos dx. Applying Bemoull’s rule, 0FOURIER SERIES OF ARBITRARY PERIOD 55 nox TEX 5 sin cos a=5 =(1)+- at (nm/i) (nn? ; 0 2 P 2! => | cos FS (eos m1), since sin nin = 0 = sin 0 nw nme -21 | a | Bye ‘Thus the required Fourier series is given by bo -2y n| nix f(xy=5+ bd wag TOV j cosy =1 To deduce the series we shall put x = 0 which gives f(x) = 0 and the Fourier series becomes, 22. Obtain the Fourier series to represent 2 in -1> Here period of f(x) = 1-(-1)=2 4 2=2 or l= ‘The Fourier series of f(x) having period 2 is given by i ae fleyegt E a,608 (ner) + x 6, sin( mex ) Since the interval is (—1, 1) let us check the given function for even or odd nature. f(x) =x-0 © f(-x) (=x)? = -x-x? which is neither equal to f(x) nor equal to — f(x). So f( ) is neither even nor odd. We shall find the Fourier coefficients by Euler’s formulae.FOURIER SERIES 1 L = ffx) de = J (r-?) de -1 a, = 4 LPC) cos (uns) dx =A 1 a, = [ (xx) cos (me) dx, Applying Bernoulli's rule, 1 a af (ro aty SAU) _ (yay) — SUM) 4 (2) _ sin(une) nl an we wre : \ | , since sin nm = 0 (1-28) cos (mx) = 2,2 | wr A 1 —4dcosnm _ —4(¢-1)" 35 [-cosrm—3eosnm| = “5 5 = ait we nn 4(-1)"*! 4,=— " we = si A COS (TX 1 in(nms) , (4) costume) = cos ( mx |x COS). cy gy me nm ne wr \ “1 72 "1 b= 0-(-2cos nm) 3 (cos HM — Cos nt) = (#1) nnn wre nm 2 yr pager "nm Thus the required Fourier series is given by 42-171 2 * (-179"1 +55 C1 cosune +? 5, Ce sin une wy ” x4 "FOURIER SERIES OF ARBITRARY PERIOD 57 23. Draw the graph of the function C JuxinOSx> Graph of f(x) = y f(x) = mx or y = xx isaline passing through the origin in [0, 1] f(x) = n(2-x) or y= n(2-x) or arty = 2m or 5 + a y 27 2n - in [1,2], which isa straight line passing through (2, 0) and (0, 2m). The graph is as follows. (2,0) 0100, 0), 0) Here f(x) is defined in [0, 2] and period of f(x) = 2-0 = 2 W=2o0r/=1 ‘The Fourier series of f(x) having period 2 is given by % = . f(x) = = + Ea, c0s (nas) +E 6, sin (am) 1 2 2 % bP fords = fpoo as + [fen a 0 0 1 1 2 J mx dx + J(2—x) de 0 1: FOURIER SERIES \ j=" ay/2 = 1/2 si a, =f J FC) cos (imme) dx D 1 a = J f(x) cos (ms) de J f(x) cos Cammy dy 0 1 1 2 = J mvcos(nme) det J x(2—x) cos (nme) dx 0 1 L 2 _ on) [ xeos (nme) dx + J (2-2) cos (rmx) dx | 0 1 j Applying Bemoullis rule to both the integrals | Si “ ns gz a enllx sin(nmx) _ 608 (1) ] i, 8 \l nT we | ° oi F cany SOD A 1)- cos (un) | | | at | | | 2 | [cosme], -[ oom | | since’ sin nme = 0 = sin 0 ae a 1 — (cos nim — cos 0 cos 2nn+cos nm). But cos 2nt = 1 = cos0 a = 2 (-242 cosnm)FOURIER SERIES OF ARBITRARY PERIOD 59 -2 4 a | 1-(-1)"} ne! ly, be T J F(x) sin (ime) dx 6 1 2 J F(x) sin Gime) aes J fe) sin Gm) dx 0 1 J mx sinQumx) det x(2-x) sin (une) dr 0 1 an = cos (HT) ay. 7x) | nm 0 2, +| (deoxy. THOM (ay. =sin (mx) ] I | nn wm 1 | 1) —x| d 2 1 a iL* cos (imx ) | *[ (2-8) cos (mr) | i = — (cos nm-0)+(0-cosun)| = 6 cos (iimx ) 1-(4+1) =0 if n is even , ~c-atyr el But 1-(-ty ]l-(-1) = 2 ifn is odd Hence f(x) = 2-4 2 ae m n=1,3,5, dl (eosme: wosGnx mx” thus f(x) = z . a 7 cos ina 7 08 Bs if | m4 3 5 J Now putting x = 0 wehave f(x) = 0 since f(x) = mx in [0,1]60 FOURIER SERIES The Fourier series becomes Aliter : (Lising the concept of even and odd finctions) f(%)is defined in (0, 2) which is of the form (0, 2!). In the given f(x) if @ (x) = me and w(x) = n(2-x), (Q-x) = O(2-N) = R(2-N) = WX) f(x) isevenin(0, 2) and hence b, = 0 a= 2 J fde, a, = 0 1 1 i J f(x) cos (nmx dx 0 =2 f nxdx = 1 ,on integration. 0 1 ai=2 j mx cos (mx ) dx 0 i 2n fF xcos (ame )de = = {4-(-1) muir Hence deduce that >> f(x) is defined in the interval { ~3/2, 3/2) 3 Bed ort period of f(x) = 3/2 (> 3/2) We shall check f(x) for even or odd nature.FOURIER SERIES OF ARBITRARY PERIOD 61 4x dx If O(X)=LE DZ, O(N) = 1- 3B RWG) f(x) is aneven function. Consequently &, = 0 The Fourier series of f(a) having period 3 is given by Peas om TX . - HTX 3/2 ] +E hysin (7) ( Qnmx 3 4, cos Lee, f(x)= Since f(x) iseven we have, 4 fox) de, since ay = 7 FF) ax 0 ifs 2277 4 /($-2-2)-0] 0 3h (2 3 ay/2 = 3/2 ¥ < Fs a, = a f fx) cos | 2a ) av since 4, = ; i fx) cos -_ dx 0 \ ; 0 ; anne) a = 1 S| de. Applying Bemoull’s rule sin 227 vow 2x | : 4 = a pa) 3 | a= He - la | "3 [ 3} nna | 5) (annay* 45 2 > “= 4 (cosmm-1) =3° 53 | cos smn 3 an? wae a, = {4-(-1)"| or a, = 8/7 where 1 = 1, 3, 5, u 22 n = Se aye a ee62 FOURIER SERIES Thus the required Fourier series is gi 1 f(x) = 5 s 3 cos ( 2arex | & w=1,3,5, a vy = 0 weget (2) = 1. The Fourier series becomes 1 8 25. Obinin the Fourier series for f(x) >> The period of f(x) = 2-0 ‘The relevant Fourier series is given by f(a) = 3 + La, cosnme+ ¥ b, sinane 1 1 2 2 2 a, = J Fox) dx, a, = J F(x) cosumeds, b, = [ f(x) sinned n 6 5 (In vach of the above integrals 1/1 = 1/1 = 1) I iys1- a,/2 = (7-1) / 26 2 a, = Jo cos my de é Wehave fe cos bedx = ——5 (aos bx +b sin by ) ath —z (—cos nix + nm sini ) | 12 ‘cosine |, sin 2m" 0 = sin 0 | cos 2un—1| = LtatnFOURIER SERIES OF ARBITRARY PERIOD ax Wehave [ ¢™ sinbxdx = '— (asin bx—b cos br) +e b, .| : 305 (Sin mmx — 7 cos nox ) [law | : =m | oy ? \ = | € * cos mmx -1 itn t 0 2 nm (=1 & (e ) "2 1antaty Thus by substituting the values of aj, a,, , in (1), the Fourier series is given by Ont cosnae + 5 5 z 2 (140'n) 1 26. Final the Fourier series of the periodic function defined by f(x) = 2x17 in theinteroat O> The period of f(x) =3-Q=3 + 2153 or The Fourier series of period 2) is given by a TX oo mx f(x) = 4 + La, cos + Yb, sin fe 1 nel The relevant Fourier series is given by ie = : 0 2nax 2am fx=y+ Xa, cos aie z b, sin” wee (1) shall find Fourier coefficients from Euler's formulac*for the interval (0,3) with ce to the Fourier series (1). That is 3 3 ; 9 7 sm 5 ay = 2 fpte) ae, ay = 2 Pfr) 005 2 ate, b= 2 fpr sin as a 4 0 In each of the above integrals 1/1 = 1/(3/2) = 2/3