Scribd
Upload
8 views

Adobe Scan 24 Feb 2024

This is a project work of a paper called Numerical analysis and the topic is Algebraic and transcendental equations.

Uploaded by

Sübhãmöy Bhãttåçhäryã
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
8 views

Adobe Scan 24 Feb 2024

This is a project work of a paper called Numerical analysis and the topic is Algebraic and transcendental equations.

Uploaded by

Sübhãmöy Bhãttåçhäryã
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 27
6.Some comparison. 2 - —_—__—__——_ Seanned with CamScanner Page-5 INTRODUCTION In this chapter we shall discuss some numerical methods for solving algebraic and transcendental equations. The equation f (x) = 0 is said to be algebraic if f (x) is purely a polynomial in x. If f (x) contains some other functions, namely, Trigonometric, Logarithmic, Exponential, etc., then the equation f (x) = 0 is called a Transcendental Equation. The equations x3 - 7x + 8 =0 and x4 + 4x3 + 7x2 + 6x+3=0 are algebraic. The equations 3 tan 3x = 3x + 1, x -2 sin x = 0 and ex = 4x are transcendental. Algebraically, the real number ais called the real root (or zero of the function f (x)) of the equation f (x) = Oif and only if f(a) = 0 and geometrically the real root of an equation f (x) =0 is the value of x where the graph of f (x) meets the x-axis in rectangular coordinate system. We will assume that the equation f(x) =0 - se eee (1) has only isolated roots, that is for each root of the equation there is a neighbourhood which does not contain any other roots of the equation. Approximately the isolated roots of the equation (1) has two stages. 1. Isolating the roots that is finding the smallest possible interval (a, b) containing one and only one root of the equation (1). 2. Improving the values of the approximate roots to the specified degree of accuracy. Now we state a very useful theorem of mathematical analysis without proof. Seanned with CamSeanner BISECTION METHOD Consider the equation fix) = 0, refer (1) where fix) is continuous on (a, 6) and fla) f(b) < 0. In order to find a root of (1) lying in the interval (a, b). We shall determine a very small interval (ay. by) (by graphical method) in which flay) flby) (by ~ ay). yee 2 If flx,) = 0 then x, is the root of flx) = 0. If flx,) # 0 and flx,) fla,) <0, then the root lies in (ox, )- In which case we rename the fnerval as (2,6), olberwise (r, by) is renamed as (a, b,) where ~ by = Felty ~ a0). Proceeding in this manner, we find Se ee inet 2 Pe a oe ee am ear ea ey, P) Where b, -a, =~ 4)s since the left end points a), a3. ...s ayy --» form a monotonic non-decreasing bounded sequence, and the right end points by, by, By, ..-+ by. --- form a monotonic non-increasing bounded sequence, then | there is a common limit c= fim a. Jim, such that f(c) = 0 which means that c is a root of equation (1). The bisection method is well suited to electronic computers. The method may be conveniently | used in rough approximations of the root of the given equation. The bisection method is a simple } but slowly convergent method. ioe a ie eee on Convergence of Bisection Method The rate of convergence of the Bisection method is linear and slow but it is guaranteed to converge if function is real and continuous in an interval bounded by given two initial guess. Accuracy of bisection method is very good and this method is more reliable than other open methods like Secant, Newton Raphson method etc. Despite being slower to converge, accuracy of this method increases as number of iterations increases. In Bisection method, error is reduced by factor of ¥% after each iterations, so we can write: @41/@, = 1/2. Which gives: Cura = Cn/2 Or ener = 0.5 e, —(1) Here en+1 is error at n+1th iteration and en is error at nth iteration. From equation (1) it can be concluded that, error at n+1th iteration is linearly related to error at nth iteration Le, Cn+sa * Cn For this reason, Bisection method is said to have linear rate of convergence. , Qrannad with CamScanner Page-8 example 22 Solve the equation x3 9x + 1 = 0 for the root lying benween 2 and 3, correct to three significant figures Solution We have P-9r4+1, S2) f3) < 275 2.88 2.94 297 297 2.955 2.955 2.9475 294 2.9475 2.9438 0.017 294 2.9438 2.9419 0.016 check 9 2.9419 2.9438 2.9428 0,003 a 4 In the Sth step a,, 6, and x, . are equal up to three significant figures. We can take 2.94 as a root up to three significant figures. *. The root of x3 - 9x + 1 = 0 is 294. Example 23 Compute one root of e* ~ 3x = 0 correct 10 two decimal places. Solution Let fx) = ef = 3x ‘fU5) = 0.02, fU.6) = 0.15 f'(x) = ef =3>0 forx € [15, 1.6], only one root of f(x) = 0 lies between 1.5 and 1.6, here ay = 1.5, Sranned with CamSranner I the ath step a, b, and x, , , are equal up to two decimal places. Thus, x = 1.51 isthe root of x) = 0, ‘correct up to two decimal places. Example 24 Find the root of tan x + x = 0 up to neo decimal places, which lies benveen 2 and 2.1. Solution Let f= nse Here AQ) = -0.18, f(2.1) = 039 ‘Thus, the root lies between 2.0 and 2.1 ‘ = 2b, =21 : ADVANTAGES OF BISECTION METHOD: 1. The Bisection method is always convergent. Since the method brackets the root, the method is guaranteed to converge. 2. As iterations are conducted, the interval gets halved. So one can guarantee the error in the solution Of the equation. + DISADVANTAGES OF BISECTION METHOD: Biggest dis-advantage is the slow convergence rate. Typically bisection is used to get an initial estimate for such faster methods such as Newton-R that requires an initial estimate. There is also the inability to ‘t multiple roots. The Iteration Method Suppose we have an equation Sx) = 0 whose roots are to be determined. The equation (1) can be expressed as x= fi. putting x = xq in RHS. of (2) we get the first approximation 1 = O(s,)- The successive approximations are then given by ame where the sequence of approximations -x,, x. .... x, always converge to the root of x=6(x) and it can be shown that if |9’(x)|<1 when x is sufficiently close to the exact value c of the root and x, 3c as ne. Faye tZ 4.4.1 Estimation of error Let € be an exact root of the equation x = 4(x) and zn41 = $(tn) Therefore, le —anl = 16(€) — O(n—1)] = 16 - 2-1 |o'(c)],¢ lies between 2,1 and NE - 2-11 YE = an +n — tn-1| < NE - 2n| + len — tn-1]- IA Il i After rearrangement, this relation becomes l pn =z,| < —Iz,, -— z»-1| < ——|21 - 20l- \é-al < Ta Ty-1| $ I il To] Let the maximum number of iterations needed to achieve the accuracy ¢ be N(¢). Thus from 1X —Ir, - | S € Teal This gives e(I-t) 10g fey=zo N(e) > ace For! < $ the estimation of the error is given by the following simple form: l€-=nl $ [tn - tn-11- Order of convergence The convergence of an iteration method depends on the suitable choice of the iteration function (z) and zo, the initinl guess. Let x,, converges to the exact root §, so that € = 9(€). Thus rn41 —€ = Olan) — (6). Let nt =2ng1 — § Note that ¢/(z) #0. Then the above relation becomes Alen +9-o8 endl) + 580 + end (E) + (Eh) i.e, Engl = end (6). Ent] " Hence the order of convergence of iteration iethod is linear, Scanned with CamScanner Example Find th 1e root 7 Asn een of x7 + x1 = 0 by iteration method, given that root {i X= 1 is the approximate value of the root lies near 1 B+x-1 0 can be put in the form 1 x= lex? such that a(x) = 1 eee, 1+ x? a ea (x)= —— (1+ xy at x = 1; we have le@=|e@)= 2 By (1+ ry so the iteration method can be applied. 9 =O(%) => l 05 1 i+ 1+P 1 ‘The root of the given equation is 0.61 after three iterations. Note: We can write the equation x° +.x~ 1 = 0 in different forms as x= and x=(1-x). Example Find a real root of cos x = 3x — I, correct to three decimal places. Solution We have f(x) =cosx -3x+1=0 (0) = cos0-0+1=2>0 Scanned with CamSeanner i an s(3)=cosk Meta ieo . A root of fe) = 0 tes tween O and ‘The given equation can be written as ee Gl cons |. Here Advantages:— *They are spaced evenly with the range of values. *The accuracy of the result is dependable. Disadvantages :— “It has a fixed rate of convergence, so it is slower than the other methods. ‘It requires a sta a ie Page 16 : ‘Newton Raphson Method: : the hewlon= Raphaen method jaa powerful ang elegant method to find the voot of an equatian, bet the equation he f(x) =O, Let xO be an approximate value of the desired root and b he the small correction to it ao that () Je the root of the equation (x)= 0 fle) = OS morh) =o Rapanding (Oxon) by Taylor's theorem, we get fom eh) = fix) WEL HO) RPC 4 =0 Ainge his small, negleot ha and higher order of h, to get {( Ho) +H P(x0) = @) On iaiiion 2) 111), We Wel the first approximation to the required root ax 0 Hieeeasive approximations can be written as fron - Aa] (0 [nem PE fu,) © The iF it converges gives the root, The Newton = Raphaon notrd ls slo lnkawa. ge ee ee tiene teil aap ) Note 11 This method falls If1'Gx) 0. Page-18 Newton Raphson method has a second order of quadratic convergence. According to Newton Raphson method: £(xn) Aah Tt ae Fie Suppose x, differs from the root abya small quantity €, so that Xp = a+ €, and Xn+1 = A+ €ns1 Now the above equation will become e+ enw = a+ en — fete F(a+éen) Snel & Se agora _ By Taylor’s expansion F(a)+enf'(a)+ze2 flat... F'(Q@)+ Enf'(a)+... enf'(a)+srer flat... f'(a)+ Enf'(a)+... Neglecting third and higher powers of e, ee e2 f(a) 2 [fi(@)+ enf'(@)] This shows that the subsequent error at each step Is proportional to the square of the previous error and as such the convergence is quadratic. Page-19 Remarks: « Newton-Raphson method can be used for solving both algebraic and transcendental equations and it can also be used when roots are complex. s Initial approximation Xp can be taken randomly in the interval [a,b], such that f(a). f(b) <0 * Newton-Raphson method has quadratic convergence, but in case of bad choice of Xo (the initial guess), Newton- Raphson method may fail to converge «© This method is useful in case of large value of f’(%q) i.e. when graph of f(x) while crossing x -axis is nearly vertical Example Use Newton-Raphson method to find a root of the equation x9 —5x+3= 0 orrect to three decimal places. Solution: f(x) =x? —5x+3 => fix) 23x? —-5 Here f(0) = Sard sU) = -1 2 FOF <0 Also f(x) is contiauons in 10,3), + atleast ene 1900 ox.sts in [0,1] nitial approximation: Let initial approximation (xo) in the interval [0,1] be 0.8 By Newton-Raphson method %,41 = *n ~ oe mn First approximation: =%- fo | where xq = 0.8, f(0.8) = —0.488, /'(0.8) = —3.08 -0.488 rhe 0.6416 => x, =0.8- econd approximation: =%4- fea , where x1 = 0.6415, f (0.6416) = 0.0561, f'(0.6416) = —3.7650 = 0.6565 £02) where x2 = 0.6565, f (0.6565) = 0.0004, f'(0.6565) = —3.7070 a 0.0004 _ Xq = 0.6565 — 7975 — 0.6566 e a root of the equation x3 —5x +3 =0 correct to three decimal places is 0.6566 | iain x = =VB 327-2w=9 -{@=7-zB = f(@=z Hes S6)= Saat 1) =8= 96). f06) <0 ~ Advantage of Newton = Raphson Method || i, suitable for large size system ii, Itis faster, reliable &the results are accurate | iii. No.of, Iteration are less to reach : A ’ rs convergence & also | iterations are independent of the . no.of.buses. | Disadvantage of Newton — Raphson Method i, Programming logic is complex than GS Method ii. Required more memory. iii. No.of.calculation per iteration are higher than GS method Seanned with CamScanner ————— = REGULA FALSI METHOD equation fix) = 0a a rat be {wo values of x such that fla) and {b) are of opposite + equation of the chorea = flswill meet the x-axis at the same point between @ and e equi of the ¢ Joining the two points (a, a)] and [b, Rb)) is y= Fla) _ s(0)~ (0) X-a ba let a < (17) = mal in (a, b) the graph of the function can be considered as a straight line, So pl coordinate of the point of intersection of the chord joining {a, a)] and {b, b)] with the x- ais will give an approximate value of the root, So putting y = 0 in (17) we get S(a) _ £)- f(a) b-a ee ele ee eee 4 a afb) (F(a) ! ae FO=sa) sa) ~ %p (say). If fla) and fix) are of apposite signs then the root lies between a and xy otherwise it lies between x, and b. If the root lies between a and x5 then the next approximation Ao af (%o) — 20f a) 1 Fle) £) _sef(i)= bls) . #7) 1) The above method is applied repeatedly till the desired accuracy . obtained. The Geometrical interpretation of the method is as follows In Fig. 2.4, the curve y = fx) Riven A(x = a) and kx = B) euls OX at Q. The chord AB cus Gaara Lag ae a ithe actual value ofthe root whereas x = OP = ys Ite Le EE ral ti, ae a Ja) are of opposite si we apply the false position mel jerval (a, oP, (tn is $v mY a ai tid we ‘ure of accuracy. The points of intersection of the neato ae _ ha Be a ‘ ¥ tend to coincide with @ the point where the curve ye “ecessive approximate values of the root of the equation. Seanned with CamScanner neste (Fiat ms copes ee oe th te pen 2 4 3-2 ns te mak inetioms Here: fix Bart SG9) = - 0H, fax Hescn tote sont Ties tosween GS aad h- ie hs oc OS, bw E FO Ota, * OH 4 cS how JOB) = - SBT G The tors Sis roves OA a08 b, ray te tensa pg we fA Lar O59 1 ye sn Nemste 1.6 sate voi fe ope tiegs iat I AEA EE thie wsing the formasda for Masts. Ys the fas sieagt- 12 ft)» 0h, (20D: Ts te 1008 en emmenn o sak 9 mad fe ee OH Ss Riss Page-24 denote the first approximation = Hlb)= F(a) _ 2(023) - ¥-06 S()= F(a) ~~ 023 — (-06) ~ 0a3 ~2” S27) = - 0.04 The root lies between 2.7 and 3. We note that £28) = 0.05 Now ~ (2.7)(005) ~ 28(-004) _ 0247 aoe 005-(-0.04) 009. “""* since (2.74) = -0,0006 “Root lies between 2.74 and 2.8. But £2.75) = 0.0081 (2.84)(0.0081) - (2.75) 2. x, = C84)(0008) ~ (275)(-20006) _ 0073844 _ 27407, 0.0081 = (0.0009) 0.0087 £2.7407) = 0.000045, the root is < 2.7407, 2.7407) = -0,000039 . ,, ~ (27406) (0.000045) - (2.7407)(~0.000039) 002301 fogs F—e = 0.002301 0.000045 (0.000039) ~ 0000084 = 2.7392 is the required value. Example Solve the equation x tan x = —1, by Regula falsi method starting with 2.5 and 3.0 as the initial] approximations to the root. Solution. We have fe) = xtanx4 1 fla) = 2.5) = 25 tan (25) + 1 = - 0.8675 fd) = G3) = 3 tan 3 +1 = 05724 By regula falsi method, the first approximation is given by 3 f(2.5)~25 f(3) _ 3(-0.8675) ~ 2.5 (0.5724) = FQ25)-f 0.8675 - 0.5728 = 2.8012 Now, ‘fis,) = 2.8012) = 2.8012 tan (2.8012) + 1 = 0.00787 (25) 2.8012) < 0 therefore, the root lies between 2.5 and 2.8012, ‘The second approximation to the root is given by 2.8012 f (2.5) - 25 (2.8012) Sranned with CamSeanner Ge oo Ges See DS ae = Se fet eon @ he oS Sree 2 29-257 CRED S- 7es-Femey Advantages *Convergence is QUarenteed: this method is bracketing} method and itis always convergent or can be controlled: increasing number of teration always yields more accurate root Does not require derivative: this method does no \require derivative calculation. *Slow Rate of Co of Regula Falsi me page-27 -:Some comparison:- | “Comparison between Regular Falsi Method and Newton Raphson Method. Newton raphson method 1.the rate of convergence is second-order or quadratic. Regula falsi method |. The rate of convergence is superlinear 2.Formula is: 2.Formula is: San) Xnotf Xn) -Xnf(Kne1) Xned F(Xn)-€Xne1) Bet = Tn ~ “les) In this method, two initial approximations 3.The computation of function per iteration is 2. 3.The computation of function per iteration is 1. 4.approximation is very sensitive. 4 Approximation is less sensitive. 5. There is no need to find |5.There is need to find derivatives. derivatives. 6. Itis not applicable for }6..Itis applicable for finding complex, multiple, } finding complex, multiple, and nearly equal two and nearly equal two roots, roots. Scanned with CamSeanner ania saye-20 B Bisection Method | Newton Raphson method 1.In the Newton Raphson method, the rate of convergence is second-order or quadratic. section Method > of convergence is thus itis slow formula 2X, 44 2.formula le.) 2 Intl = Lp — FGn) 3.The initial approximation is |less sensitive. The initial roximation is very sensitive. In the Bisection Method, there is no need to find derivatives. 3.The initial approximation is very sensitive.In the Bisection Method, there is no need to find derivatives. In the Newton Raphson method, there is a need to find derivatives. Regula falsimethod bisection Method 1.This method faster order of } 1.In the Bisection Method, convergence than the bisection} the rate of convergence is method. linear thus it is slow. 2.formula Xn-1f(Xq)-Xnf(Xn-1) | 2.formula * Xn" Sree a, oe n+ = i i known as the 3.Itis also known as the 3.It is also ‘ ition method Bolzano method, Binary — chopping method, half Interval method Scanned with CamSeanner | Conclusion poo yeet was a great way to help myself realize some that | before, Like how mueh work | actually put yoy) assignments and how much Lactually understand The work (hat is put fa front of me, Usually we don't look ai the work after we go home because, do it at home and you have frends around you at college to help divide the work, This is how student's minds work, and actually how [think about it sometimes too, Sometimes | don't understand the lmportanoe of teachers having us do these projects that seem to take a lifetime, but then atthe end of the day when it's all over [finally grasp the concept and the idea of the whole thing and why they make us do it in the first place. Because they want us to learn the importance of what we do in class or what we have learned and make sure we don't leave this college with a miss understanding, Roannved with Qaaitownrier Reference jhave done it by myself with the help of some t yhich are given in their reference * Numericals analysis - G.shanker Rao. *Numericals methods- S.R.K lyenger,R. *Numericals analysis- Madhuman *Numericals analysis - S.A Mi *Numericals analysis - Arup Mukherje *Some Notes

You might also like