KEMBAR78
Numerical Methods | PDF
Scribd
Upload
408 views56 pages

Numerical Methods

Uploaded by

Alok Jangra
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
408 views56 pages

Numerical Methods

Uploaded by

Alok Jangra
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/ 56
~ .1. Explain bisection method. : Solution. Bisection or Bolzano method : This method is based on the repeated application of intermediate value property. Let the function f(x) be continuous between a ‘and b. For definiteness, let f(a) be —-ve and f(b) be +ve. Then the first 3 1 approximation to the root is x; = 2 (a+b). XY If fix) = 0, then ‘x, is a root of-f(x) = 0 otherwise, the root lies between a and x, or x, and b according as f(x;) is + ve or — ve. Then we bisect the interval as before and continue the process until the root is found to desired acturacy. In the above figtire}x,) is + ve so that the root lies between a and x,. Then,second approximation to the root is x2 ss 5 (a + x4). If ‘Ke2) is ‘ve; the root lies between x; and xy. Then the third approximation to the root is x3 = 5 (x1 + xg) and so on. Once the method of calculation has been.deciled. We must describe clearly the computational steps to be followedin a particular sequence. These steps constitute the algorithm of method. Q. 2. Explain the algorithm of bisection method. ~ Solution. Algorithm of Bisection Method : ‘ Step 01. Start of the program - : i Step 02. Input the variable 1, x2 for the task Step 03. Check f(x) * fw2) <0 Step 04. If yes proceed Step 05. Ifno exit and print error message Step 06. Repeat 7-11 if condition not satisfied Step 07. x0 = (v1 + x2)/2 Step 08. If Ax0) * Axl) <0 Step 09.x2=x0 © Step 10. ELSE Step 11. x1 =x0 Step 12. Condition ‘ Step 13. | (x1 -x2)/x1 |< maximum pole error of, F flx0) = 0 Step 14. Print outout Step 15. End of program Q. 3. Explain the order-of\convergience of bisection method ? Solution. Order of convergénce:of bisection method: In bisection method, the original ‘interval is dividecl into half interval in each iteration. If wetake mid-points of successive intervals to be the approximations of the root, one half of the current interval is the upper bound:to,the error. In bisectionmethod, : : 0412 0.5 e st A) i where)e; and e; , ; are the errors in the i‘ and (i + 1) iterations respectively. Comparing (1) with 0i.e., positive i i 15. .. The required root lies between 1 and Iteration 2. Let a = 2 then f(2) = 5, proved above fc) = f(1'5).= 0875, proved above and Let d= Flate)= 2 (1416)=125 d) = f(1-25) = (1-25) — (1:25) —1 i le gt ri [from (i] =- 0-297 < 0i.e., negative -. The required root-lies between 1:25 and 1:5. Iteration 3. Now d = 1:25, c.= 1-5, where — Ad) = f(1-25) = - 0-297, proved above. and. Re) = f(1-5) = 0-875, proved-above 1 1 vd i es = (1 25) = = (2-75) = 1-37: Let e= 9 era) 3 (154 1:25) 3g (2°75) 1:375 Re) = (1-375) = (1:375)3 (1-375) — 1, [from (i)] = 25996 — 1-375 S1'= 0-2246 > 0 ie, Positive :. The required root lies between.1:25 and 1-375. Iteration 4. Now d = 1-25, = 1-375, where Kd) = f(1-25) = 0-297, proved above and Re) ='375) = 0:2246, proved above Let h= AG te)= 5 (125 +1375) = 1-313 Ah) = (1313) = (1-318)? (1-313) - 1, [from (i)] ; : = 22636 - 1313-1=_0-0494 < Oie., negative he The required root lies. between 1:313 and 1-375, Iteration 5. Now h=1318 ande = 1:375, where - Rh) =-0-0494, Re) = 0-2246, proved above -. Let k= Fh+e)= Sais +1:375) = = (2-688)= 1344 Ah) = (1-344) = (1-344)8~(1-344)—1, [from (i)] = 24277 = 1:344-1= 0:0887 > O i.e, Positive -, The required root lies between 1-313 and 1344, Iteration 6, Now h = 1-318 and k = 1:344, where fh) =~ 0-0494 and f{k) = 0-0837, proved above 1 ° Let me sh+h)= Li sig W844) 10 pes yecrenn Aim) = (1:829)8 - (1-329) -- 1 = 23478 — Load — 1 = 0:0183 > 0 i.e., positive +. The required root lies between 1-313 and 1-329. Iteration 7. Now h = 1:313 and m = 1:329, where fh) = — 0-0494 and f(m) = 0:0183, proved above 1 ell Let p= gh +m)= 3 0-818 + 1:329) = 1-321 yess flp) = (1:321)3 - 1-321 - 1 = 23052-2321 =-0-0158 < 0i.e., negative <. The required root lies between. 1:321 and 1-329. Iteration 8. Now p = 1:321 and im = 1-329, where fp) = (1-321) =-0-01.58 and" fim) = f(1'329) = 0:0183, proved above | ‘ Let q=500+m)= (1921 +1429) = 1:325 Aq) = (1-325) = (13285) — (4325) — 1 = 2'3262 — 2-325 3=0:0012 > 0 ie., positive :. The required root lies between'-321 and 1°325. Iteration 9. Now p = 3-321 and‘g = 1-325, where - fe) = f(1-321) > — 0°0158, fla) = f(1-325) =0-0012 Let r= 5¢ +Q)= F132 + 1:325) = 1:323 fP2A1-323) = (1-323)8 — (1-328) = 1 = 2:3157 — 2-323 = —0-0073 < 0 ie., negative :. The required root lies between 1-323 and 1-325. _Ffom here it is evident that upto two places of decimal the reqired root of the given equation is 1:32. »Q. 10. Apply bisection method to determine a real root of the equation f(x) = 8x3 - 2n-1=0. Solution. Given fl) = 8x3 - 2x -1=0 ci) ; AO) = 8(0)8 - 2(0) — 1 = — 1 (negative) and f) = 8(1)8 - 2(1) — 1 = 5 (positive) ++ f(0) and f(1) are of opposite signs, so at least one root of the given equation lies between 0 and 1, : Iteration 1. Let a=0,b=1 1 1 Let e= g(a+b)= 7 0+1)=05 Ae) = R0°5) = 8(0'5)° — 2(0°5) — 3, {from (jy *= 8(0:125) -1-1=-1<0i.e., negative -. The required root lies between 0-5 and 1. Iteration 2. Here b = 1 and c =0°5 fic) = (0-5) — 1 and f(b) = f(1)= 5, proved above 1 Let d= FO +0)= 5 (1405)=075 Ad) = f(0-75) = 8(0-75)3— 2(0°75)— 1, . [from Gi) (0-4219) — 1-5-1 =3-3752-2:5 G = 0:8752 > 0 ie., positive _ *. The required root lies between 0:5 and 0-75. Iteration 3. Now c = 0:5 andd = 0:75, where Ae) = X0-5) =— 1 and Ad) = K(0-75) = 0°8752 Let e= S(c+d)= 505 + 0°75) = 0-625 Re) = (0-625) = 8(0-625)3 — 3(0-625) — 1, [from (i)] = 8(0-244) -— 1.25% T= 1-952 — 2-95 =— 0-298 < 0.) negative A -.. The required root lies between 0:625 and 0-75. Iteration 4. Now e = 0:625, d = 0:75, where Rd) = f(0"75) = 0-8752 and fle) = (0-625) =~ 0-298 8; 1 Let ‘he ge +d)= g (0625 + 0°75) = 0-688 Kh) = f(0-688) = 8(0-688)3 — 2(0-688) — 1, [from (i)] = 8(0°3257) - 1-376 -1= 2-6056 — 2:376 . = 0-2296 > 0i.e., Positive .. The required root lies between 0-625 and 0-688. , Iteration 5. Now e = 0-625, h = 0-688, where Soa fle) = K0-625) +298 Rh) = K0-688) = 0-2296, Proved above fi 1 : Let k= ie +h)= a (0:625 + 0-688) = 0-657 Rk) = (0-657) = 8(0-657)3 — 2(0-657)-1, [from @) - =8(0-2836) — 1-314 -1 = 2-268 —2:314 =-0:0452<0 . ; «. The required root lies between.0:657 and 0-688, Iteration 6. Now k = 0-657, h = 0:688, where Rk) = (0657) =— 0-0452 and Rh) = (0-688) = 0-2296 i 4 Let m= 9 (kh+h)= 3 (0-657 + 0'688) = 0-673 fm) = f(0'673) = 8(0-673)3 — 2(0'673)-1, [from @) = 8(0'3048)—1:346—1 = 2:4384 2-346 =0-0924 >0 -. The required root lies between 0-657 and 0-673. Iteration 7. Now k = 0-657, m = 0-673, where lk) = A0-657) = — 0-0452, flm) = (0-673) = 00924 c 1 1 Let p= 5 (k+m)= 5 (0-657 + 0673) = 0-665 fip) = A0-665) = 8(0°665)3 - 2(0-665)—1, [from @l = 8(0:2941)—1-33- 1 = 2:3528 2:33 =0;0228 > 0 :. The required root lies between 0-657 and 0-665: " Iteration 8. Now k = 0:657, p = 0'665, wher@\ fle) = 0-657) = — 0-0452, Alp)= (0665) = 0-0228 1 1 Let q= 5b +p)= 5 (0657 + 0-665) = 0-661 » fq) = (0-661) = 8(0:661) —2(0-661)— 1, [from @)] - $8(2-888)-1:322-1=2'3104—2:322=— 0:0206<0 -. The required. root lies between 0663 and 0-665. From here it is evident that upto two places of decimal the required root of the given equation is 0°66. Q. 11. Explain false-position or Regula-Falsi method ? Solutions; Method of false position: or Regula-Falsi ~ methods The bisection method guarantees that the iterative process will converge. It is however slow. Thus attempts have been made to spéed up bisection method retaining its guaranteed convergence. A method to doing this is called the method of false position. It is sometimes known as method of linear interpolation. Ao: {Xo}. 1s 1s the oldest method for finding the real roots o¢ a ruherisil ‘dation and closely resembles the bicection methog, In this method, we choose two points x» and x; such that Rx) and f(x;) are of opposite signs. Since the graph of. y= f(x) crosses the X-axis between these two points, a root must lie in between these points. Consequently, — flxo) flx;)< 0 : Equation of the chord joining points (xo, flo)) and (x1, Alc) ig ) Flea) ¥~Reg)= FED= L003) -The method consists in replacing the curve AB by means of the chord AB and taking the Point of intersection of the-chord with X- axis as an approximation of the root. So the abscissa of the point where chord eutg Y = 0 is given by Exy— “1% ei X2=X Fey) Fla) (1) which is an approximation to the root, : Tf now flxo) and Axa) are of opposite signs, then the root lies between x) and %2. So replacing 2rby x2 in (1), we obtain the next approximation x3. However, the root could as well lie between x, and x» then we find 3 actordingly. i This procedures Tepeated till the root is found to the desired accuracy. \ . Q. 12. Writeian algorithm for Regula-Falsi method. Solution, \” Algorithm _ Step 01. Step of the program, | : Step 02. Input the variables x0, «1, e, n for the task. Step 03. /0 = Alxo) . Step 04, f1 = flx1) Step 05. for i = 1 and repeat ifi<=n Step 06, x2 = (x0 fi-x1 PO)KF1-f0) Step 07. 2 = x2 ' Step 08, if | 2 | <=e Step 09. Print “convergent”, x2; f2 Step 10. If sign (72)! = sign (fo) . Step 11. x1 = x2 & fl =f2 Step 12. else : Step 13. x0 = x2 & f0= 72 Step 14. End loop Step 16. Print output Step 16. End of program. Q. 13. Find a real root of the equation 3x + sinx-e*=0 by the method of false position correct to four decimal places. Also write its program in ‘C’ language. Solution. Let f(x) =8x + sin x -e*=0 . 0.3, 0.154 i.e., — ve and A0.4) = 0.0975 i.e., + ve «. The root lies between 0:3 and 0.4. Using Regula-Falsi method, xy — X Xp = xq — =A — fx) f(x) — F(x) (0.4) - (0.8) (0.0975) — (— 0.154), [-. xo = 0/3"and x, = 0.4 (let)] 0.1x 0,154 = (0.3) + (*ax2te8) =0,3612 Now, fxg) = (0.3612) = 0.0019 = positive Hence, the root lies between 0,3 and 0.3612. (x2 — x9) Flaad= flay =(0.3)- (0.154) Now, again, %3=X%9- [Replacing x, by xq] (0.3612) — (0.3) = (0.3)— 10019) =(— 0.154) (- 0.154) (ob) = 0.3604 = (0.3) + ae >: 0.1559 Now, fxs) = (0.3604) = — 0.00005 = negative ~. The root lies between 0.3604 and 0.3612. é Now, again, aie | aie None aed {ect [Replacing xo by xs] (0.8612 - 0.8604 | -0.00008) | (0.0019) —(— 0.00005) 0.0008 0.00195. = (0.8604) — [ | ws = 0.3604'+ ( ).00005) = 0.96042 VMAce Xg ANG x4 are approximately the same hence the Tequire; Teal root is 0.3604 correct to four decimal places. ©. Q. 14, Find the root of the equation xe* = cos x in the ‘interval (0, 1) using Regula-Falsi method correct to fou, decimal places, Write its computer program in ‘C’ language Solution, Let ' fx) = cosx — xe* = 0 so that AO) = 1, f1) = cos 1-e = — 2.17798 ie., the root lies between 0 and 1, By Regula-Falsi method, (% = x) F(x) = F(x) “Xp = 29 — Axo) 1-0- =0- ———_ (1) = 0:31467 8 = 3.17798 ree Now, Rex) = 0.31467) = 0.51987 - Le., the root lies between 0.31487 ‘and 1. : (430.31487) ; : . =0. ¢_6.e00 7 DD 5 Again, x3 =0.31487 © 2.17798 —0.51907 (0-51987) = 0,44673 3 Now,” fix) = 0.20856 «. The root lies between 0.44673 and 1. Repeating this process, X19 = 0.51775 corrected as 0.5177 upto 4 decimal Places. Q. 15. Find'a Feal root of the equation x3 -2x-5=0by the method of false’position correct to three decimal places, Solution. Let Rex) = 3 — 9 5 x sothat . f2)=—1and f3)= 16 : i.e., A root lies between 2 and 3. Using Regula-Falsi method, ~ (1 = 49) F(x)= Fx) (8-2) =2- Geen 1) = 2.0588 Now Reg) = (2.0588) = ~ 0.3908 The root lies between 2.0588 and 3; XQ =X — Kx) Now, again x, = 2.0588 -( o222.0588 Je 0.3908) = 2.0813 16 +0.3908 ftvepeating tis process, the successive approximation are X4 = 2.0362.....xg = 2.0948 etc, Hence, the root is 2.094 correct to three decimal places. Q. 16. Using the method of false position, find the root of equation x5 ~'x4 - x3 - 1 = 0 upto four decimal places. Solution. Let x)= x6 -xt—x3-1 “y .4)=- 0.056 A141) = 0.102 > Hence the root lies between 1.4 and 1.41. Using method of false position, xy nan (as Feat =14- ( 141-14 0.102 + 0.056 (Let xg 1.4 and x; =1.41) Je (— 00 age 0.01 -14+ (28 + (6.056) = 1.4035 Now, Ap). = — 0.0016, ee Hence the root.lies between 1.4035 and 1.41. Using the method of false position, x} . xp — [aztohe fixe) (Replacing xp by x2) 1.41-=1.4035 0.102+ 0.0016 0.0065 0.1036 Now, fxs) = — 0.00003 negative Hence, the root liés between 1.4036 and 1.41... Using the method of false position, i % 1.41-1.4036 0.102 + 0.00003 “ 1.4086 -( (0.0016) = 1.4035 + (eevee Joo 0016) = 1. 4036 = 1.4036 + (2 )c.00008) 0.0064 5} 0 00003) = 1.4036 0.1020 Since x3 and «4 are approximately the same upto four places g decimal, hence the required root of the given equation i is 1.4036, Q. 17. Find areal root of the equation x logo #= 12 by Regula-Falsi method correct to four decimal places. Solution, Let fix) = x logyx —1.2 ' d \ 21 4036+ (5 Since f2.74) = — 0005634 fl2.741) =.0003087 Let xp = 2°74 Xy = 2-741 Hence, the root-lies between 2.74 and 2. RN Using method of false position, and. Kp an Or? Pts bes sea (Let x9 = 2.74 and x, =.2.741) 2 td 2-741 - 2.74 : = pv ead 2.74 “a. [-0008087 — 0005684) [| (--0005634) YY. 001 99= 2.744) ———_ NS +(, aoe] (J 0005634) =2,740646027 \ “f%2) = — .00000006016 i.e.,— ve «Hence the root lies between 2.740646027 and 2,741, \ " ~ Using method of false position, Now, = { Asa } hi Repl b Xg=X_- Fe) =F) Ap). (Replacing x, by x9) 2.741 — 2.740646027, ie fee & eae + DODNUOETTS) i : x (+,00000006016) = 2,740646096 Since x2 and x3 agree upto seven decimal places, the required ront:correct to four decimal ninonc {4.9 FANG method. e, so that Q. 21. Explain Newton-Raphson method. Solution. This method is generally used to improve the Tesult obtained by one of the previous methods. Let x» be an approximate root of fix) = 0 and Let x, = xq + be the correct root so that fix;) = 0, Expanding f(xp + h) by Taylor’s series, we get i 2 ‘ fix) + hf'(xo) + cai f"Go) +. Since, h is small, neglecting h? and highér powers of h, we get Axo) + hf'(xp) = 0 or h'= — Leo - +) \ Fo) A better approximation than xo is therefore, given by x;, where \ fo) 4 = eh 4 Lo? 1= Xo FG) Successive approximation are given by x2, X3, ..+) Xp 41, Where . 1 F(%n) : Br 1=Sn~ Fy (n= 0,4...) 2), ‘ which isthe Newton-Raphson formula, Q. 22. Explain order of convergence of Newton-Raphson Solution. Suppose x, differ from the root aby asmall quantity Xn 0+ ey aNd ty r= A+ ey yy =x. 2H login tn 1.2 48429x, +1.2 A) ™™ logo Xn +. 43429 — logy Xp + 43429 Put n = 0, the first approximation is _ :43429% +1.2 © logy 2+.43429 Similar ly, putting n = 1, 2, 3, 4 in (1), we get tg = 2.741, xg = 2.74064, x4 = 2.74065, 25 = 2.74065 Clearly, 4=%5 Hence, the required root is 2.74065 correct to\five decimal places. Q. 25. Evaluate ‘V12 to four decimal places by Newton’s iterative method. C Solution. Let x= J12 so,thatx?-12=0 » Take flx) = x2 — 12, Newton’siteration formula gives, =281 Cm =2) 1 (Xn) Xn +1 =n rs ) oe n aw Sage (12 S aes eae (« " 2) @) Now,isince. -f(3) = — 3 negative f(4) = 4 positive -. The root of (1) lies between 3 and 4. , Take xo = 3.5, (2) gives, = 38.4641 Since, x, = xg upto 4 decimal places. Sowehave 12 =3.4641. Q. 27. Find a real root of the equation « =e-* using th, Newton-Raphson method. - Solution. We have Rx) = xe* — 1 then fe) = (1+ x)et Let. xo=1 then Os epee (=) = (193) = 0.6839397 i 2e 2\K Now Rey) = 0.3553424 and f'@1) = 3.337012 cm, 0:3553424 : a # =0. Dassaee = 0.577454! so-that Xe 0. 883935 3.337012 0.57745: 5 Proceeding in this way, we obtain ee z E x3 = 0.5672297, %4 = 0.5671433. Hence required root is 0.5671 correct to 4 decimal places. 'Q. 28. Find to four places of decimal, the smallest root of the equation e-*>sin x, : Solution. The piven equation is - Rx) = e-*— sin x =0 “ ie ven in +$in xe so that Rng 1 tn Trias eos . Take xp = .6, then (FEAL UM NAL ocd y %1 = 58848, xo\= 588559 Cher Hence desired value of the root is 0.5885, Q. 1. What do you understand by interpolation and extrapolation? Solution. According to Theile, ‘Interpolation is the art of reading between the lines of the table.’ It also means insertion or filling up intermediate terms of the series, Suppose we are given the following values of y = fx) for aset of values of x : x: Xo. x. Xo, ee Xn vy: Yo M1 2 us Yn Thus, the process of finding the value of y corresponding to any value of x = x;, between xp and x, is called interpolation. Hence interpolation is the technique of estimating the value of a function for any intermediate value of the’independent variable while the process of computing the value,of thé function outside the ~ given range is called-extrapolation, . : Q. 2. Prove that: A=E-VorE=14+A. Solution. We know that,“ eye / Ad's = b+ Ie = Ey — Ie = (EB - Vy > A=E+1 ‘ or « HE T+ A. : Q. 3. Prove that, V =1-E-1. Solution. Wye = V2 —-Jz-h=I2—- EY a V=1-E-1. Q. 4»Prove that : 5 = E}? - E-1”, Solution. Byn= Yh -I, kh = EM yy Ey, : 2 es ato "= (EY E-V2)y, 6=EV2_ R-V2, 1 Q. 5. Prove that : p= 3 (EV2 4. B- 12), rary Bi 1 Solution, Wy = 2(7+4 ty S 4) = ge +E yy, > “pe 5 + E-1?), and point out theivalues of A%y,(, Atys. Q. 6. Prove that : A= EV = VE = 5E™. Solution, 2(Vy.) = BQz- Yen) =Ix+h—Ie = We > Ev=A VE ys) = Voth = 24h ~ Ix = AV > VE=A | i SEM y= gy 4 = Ieah Ie = We, xe 2 >. SEY = A, Q. 7. Prove that: E = e!D, Solution. - Ef{x) = flx +A) asp = fx) + hf") + eras Tai (By Taylor series) 2 . =f) +hDflx) + phe oA (ab? = [topo 022 2] erie) E=e!D onae sn 1. Q 8. Construct the forward" difference table, given that xt] & | 40 | 15\] 20°] 25°] 30 + | 9962. | 9848 |'9659 | 9397 | 9063 | seco |: Solution. Forward difference table is-as follows : = a RN To By Ay Aty 5 9962 114 | 10 9848 “ 18 : -189 ‘ 2 15 9659 : = 262 1 20 9397 2 ~ 334 3 25 9063 -69 ~ 408 30 8660 : From the table, A*y9 = — 73 and Aty, = Q.9.1fy = x9 +4226 +1, calculate values of y for* = 0,1, 2, 8, 4, 6 and form the difference table. Find the value of y at ve by extending the table and verify that the same value is obtained by substitution. ; Solution. For x =0,y=1; x=5,y=141 "+, Third differences are constant. p A®’yg=6 => Ary, — A*yg = 6 =>. Ay 26=6 > A’y,= 32 Now (A%y 4342 = As — Ayg = 82 => Ays — 68 = 82 => Ays = 100 Further, Ays = 100 ¥6=Y6= 100 => Yye—-141= 100 Yo= 241 Verification :y(6) = (6)3 + (6)?-2(6) + 1 = 241. Hence verified. Q. 10. Construct a backward difference table for y=log x given that: and find values of V3 log 40 and ¥ log. 50: Solution. Backward difference tabke is : x y WN Wy Vay -[ Vey 10 1 a ‘ 0.3010. 20 1.3010..|_-\). -| 0.1249 0.1761 | 30 VA711 — 0.0511 — 0.0508} © | 0.1250, 0.0230 40.) 11,6021 | = 0.0281 ne 0.0969 50 _|_1.6990 From the table, V3 log 40 = 0.0738 and V4 Ing KN =— 0 nkna As only three entries 79, 2,4 are given, the function y Can he represented by a second degree polynomial. Ee A%yq = 0 and A’y; = 0 > 3y1 +g = 9 and y; + By3 = 3.6 Solving these, we get ¥1 = 2.925, yo = 0.225. Q. 14. Find the missing value of the following data ; eee pe 2 | 3 4 5 fw: | 7 | | as [a 37 Solution. Since four values are known, assume all the fourth order difference as zero. Since one wee is unknown, We assume Aty,= > (E-1)4y,= > Gham eer ans ty ao 3 . Bate Ys — 494 + Cys — 4yo + y= 0 Ceh=1) > > = 37 —4(21) + 6(13) — 4y 4°70 38=4)y= 0 ¥2= 9.5 Hence, the required miding value is 9.5. Q. 15. Find the value of sin 52° from the given table: @° 45% 50° - 55° 60° sin 6° 0.7071 0.7660 0.8192 0.8660 Solution, a=45°,h=5,x=52 7 seh . EE 5 4 Difference table is y, Differences fe 104 y tay ~] 10!A% O*ASy 45° 7071 589 50° 7660 57) : 532 ° N =7 55° 8192 -64 468 60° 8660 By forward difference formula, ao > u(u—1)(u— 2) , flathu)=fla)+u Afla) + 22 fa) + a A3fla) mas » > ee 104 fla) + 104 u rab +104 A? fla) 2 + 10 Pata ae 7 fa) = 10452) = 108145) + (VALIOAA 45) + ce} 104 A? (45) 1.4)(1.4—1)(1.4—2 + J 104 A3 (45) Sor + (2.4) 689)+ “2 C57 1.4).4)-.6 EME Oy = 7880 52) = .7880 Hence sin 52° = 0.7880. > 105 (58.75) = 176900.588 . 58.75) = log 58.75 = 176900.588 x 10-® = 1.76900588 Hence log 5875 = 1.76900588 (-.« Mantissa remain the same) Q.27. Apply a central difference formula to obtain f(32) iven that : : (25) = 0.2707, f(35) = 0.3386 {(30) = 0.3027, f(40) = 0.3794. Solution. Here athu=382andh=5 aie origin at 30. a= 80 then u = 0. 4 ierward difference table is uw x Fe) Af | A%fe) Ax) -1 25 | .2707 ; : / 032 0 -{'30-° |) £8024 .0039 : & 10359 1s.[25.85 ‘3386 0049 , .0408 aod 40. 3794 Applying Gauss forward aitirence formula, we have flu) = f0)'+ uAfO) + 5 wee D nf) + eee Re D ; . AY4-1 A.4) =.3027 +. (41.0859) + (Aha (.0039) (1.4).4)04 1 + eee (.0010) = 0.316536. Difference table is : Gauss’ forward difference formula i’ (ut ene 1) - uurdDs,* fu) = 0) + udfl0) + ay wr" 1) + A%f- 1) (utIuu-Du = 2) Rear 4 Torch Atf{—2 = Lae 3-2)+ ee ? (x — 2)(x- 3-4) pee She eel 6 ae (x = 2)(x = 8)(x — 4)(x— 5) i 24 (8) (16 4 =1-20+6—-2x2+ 140-244 gO 9x? + 26x - 24) + 2G4- 143 + 71x? — 154% + 124 : 100 | F(x) = oA 8x3 + Zee 56x +31 Q. 55. Using Lagrange’s interpolation formula, Pen) rom the following table: ef 6 peer 78 eat | | y | 12] is] 14 | 16.) Solution. Here x9 = 5, x; = 6, x2 = 9, x3 ; fxo):= 12, fix) = 18, Rea) 3 = 14 fog) = 16 Lagrange’ 's formula is _ = xx — x2 (4-13) . i (xo — %1)(% — x2)(x ~=ys ‘ (% — x9)(x — x9 (ar arg) flocs) (a = Xp Nay ea) — 9) (% = x9 x = x1 (x — x3): > (9 — X9 ary — 201 (xq — x3) Ae) (6 = xq x — 4 x = x0) : ‘i , * (xg — Xp x3 — x4 (ag — x2) Aas) (x —6)(%— 9)(x — 11) (x —5)(% —9)(x — 11) f= Ga6y5-916-11) * (@—5\6-9N6-1n) ~ (x — 5)(x —6)(x-11) (x—5)(x-6)+(x-9) Gi—sat—@ar— 9) (16) * @-5@-69-1) 1+ i =~ 9&6) 9) @-1) + Fe B= 9) &- 1D) 7 ~ Jg 5) @-6) (eA) + F(x 5) (e- 6) (2-9) Putting x = 10, we get . 1 A110) = - 3 (10 ~ 6) (10=9) (10 - 11) 13 w jp (10-5) (10 — 9) (10 -.11) a ; 4 = Ja, (1078) (10 - 6) (10-11) + 75 (10 - 5) (10 - 6) (10-9) = 14.66666667 Hence (10) = 14.66666667. Q. 60. Use Lagrange’s interpolation formula to Tv 4 polynomial to the data: wi [-s[ 3 [1 [2]. Hence or. otherwise find the value of 7. * Solution. Here x) = 1, x1=0, x%=2, xg=3 ‘0.) = — 8, fly) = 8, flag) = 1, flvg)'= 12 Lagrange’s interpolation formula is (= % (x - x2 x= 43)~ feos (xq = x4 (x = x2 Mag — 3) feo) (x = x9 a = 9 (x — 5) Gwe (x Lag (% — x1 x= x3) = ) + Ge ap My — a Vea — xy)? (x = x9 (x = x4 (x — x2) 3 * (xg — Xo Mag — x1 xg — xg) Revs) ae OMac— 2)(x — 3) © (+1 — 2X38) = Gal 1-H 1-3) —® *o.n0-2029” (% +1)(x— 0x = 3) (x+ I(x — Ox — 2) P@rne-02-3) * @+VE-H8-3) G2) = Fale 2)(¢-8)+ $+) @-2@-3) | = H+ Dx —3) + &+ 1) xe —2) 1 Sees 1 : = = (3 5x2 + 6x) + gh be + x + 8)— G GS 2x? — Bx) ‘i vy ‘ + (xB Hx? 2x) - Q. 61. By means of Lagrange’s formula, prove tat; Gi) 99 = 0.059494) - 0301430 de Cee ya) iy eae OsU eae ORG ot: Son aoe ras catia -3,-1, 1,3, the Lagrange’s formula is : é Get Dee DnB) 5 i AD DE 3)_ ** CB4DE3-DC 3-3)” 9 C148 1-DE 1-3"! (x+3e+D@-3) | @+3\@ID@-1) * Geaaspa—a)??* @r+a@+DE-H” (x +1\(x-1)(x—3) (x+3)(x-INe> 3) Ts GAB eee Sohn He (+3) + 1-3) > \y (0 $B)(x +1)(x —1) Ci) IN aes Putting x = 0 in (1), we get , 1 ee Re OF Tg 8 ag 1 tg 1 = Te 73 = cr +9-1) - ho ~n)- 50. 1-9-3) (ii) For the arguments 0, 1, 2, 4, 5, 6, the Lagrange’s formula is (x= De >2)(x — 4)(x~ 5) —6) : %*= ONO -2K0- 40 —5\0—6) 70 (= 0)(x — 2)(% —A)(x~ 5x ~ 6): * @-00= 20-40 —8y1 6) 1 (% = Ox — lw ~ 4) — 5) — 6) * @-0@-D2-42-Ba> 6) 2 ‘(e— Oa ~ Ne 2x — 5x6) * @=04-D4= aa — Ba 64 ¥ (= Ox ~ 1% — 256 4)(x—6) . G- 06-6 = 2x6 —aye— ey (x ~ ON ~ Ia — Ne — 4) 5) 6-0\K6- HE 26 =4)(6—5) 76 D) Putting x = 3 in (1), we get > 33 = 0.05y9 — 0.3y1 + 0.75yo + 0.75y4™ 0.8y5 + 0.0576 = 0.05(yo + 6) — 0.3(y1 + y5) + 0.75(yo + 4) ' Gii) For the arguments — 5, -— 3, 3,5, the Lagrange’s formula is _ __ (%+8)(%—3)(x-5) (x + 5) — 3x — 5) (5 +8)— 5—3N- 55) 7° NEB +5\— 8-3) 8-5)” 9 (x + B)(x + 8x5) (x + B(x + 8)(x — 8) 1) (8+ 5)8#3\3- 5)? * G+5)5+3\5—3)°o Putting x = 1 inequation (1), we get 1 = 7 0.2y¥5 + 0.5y_3 + y3 — 0.895, =9g 0.3(¥5 —y_a) + 0.2(y_a—v_«). £3 i merical Glee auto Nea erica differentation : ake Process calculating the derivatives of a function with the hel lp of a given a of values of that function is called the numerical differentatio, : Q. 2. Give an introduction of the Numerical Integratio, da * ¥ sus s * hed Numerical Integration : By definition, the integral y same function f(x) between the limits a and b may be thought ofa the area, A between the curve and the X-axis. This configuration, shown in Figure and ‘is written mathematically as AsProa | oo Fig. : The integral of f(x) froma to} represented.as the ay’. _ » area under the curve, By the fundamental of calculus, we ‘know that. continuo functions have an exact analytical solution: © Using F(x) to denote the antiderivative of Kx), this solution given by en Bee fp feo ae = me) _ Faye tes 7) In practice, it is actually. Common fo; el) ‘ ‘ : rE trem difficult (if not impossible) to obtain, thus fae 2 e e ttl efor? | pumerica® approximation instead, This commonly occurs with — | gpplications that require definite integration of highly complex | hysical models or discrete samples of random data. Consequently, ymerical integration has become an indispensable tool for processing sophisticated engineering designs, It is, therefore, important to gain, an appreciation for the scope of numerical integration and its power to solve real engineering problems. In general, there are three major trade-offs to consider when choosing a particular algorithm for definite integration. The first and most important, is obviously the accuracy of a given numerical approximation. Fortunately, no integration scheme is so inaccurate that is cannot be compensated for by iterating the integration over a geater volume of data samples. Thus, the second metric to consider isthe computational time required to achieve a given‘aceuracy. This igusually measured in terms of function evaluations.or fevals, that the computer needs to perform while implementing the algorithm. Many advanced schemes also exist to provide very high degrees of accuracy with few fevals, but generally require for more time just to write the code than to simply grind-on.a more primitive version. The complexity of the algorithm is, therefore, the third major trade- off to consider. The “correct” \choice-of algorithm must then be determined by the nature of\the application, the available computational resources andthe technical experience of the designer. Q. 3. Explain Maxima and Minima of a Tabulated function. . . Ans. Maxima\and Minima : One of the great powers of calculus is in the determination of the maximum or minimum value ofa function.\Take tx) to be a function of x. Then the value of x for Which: the derivative’ of f(x) with respect to x is ‘equal to zero torresponds to a maximum, a minimum or an inflexion point of the ion fix). , For example, the height of a projectile that is fixed straight up Sgiven by the motion equation : Taking yo = 0,'a graph of the height y(t) is shown below : 1 ie IE) = Voyt - 38 dy & xu -gt=0 atm wae Q. 10. Explain Simpson’s 1/3 rule. Solution. Simpson’s one-third rule®,Put n = 2 in genend quadrature formula and neglect third andchigher differences, ta this case interval of integration will be‘from xo to x9 + 2h and there will be only three functional values'yo, y; and yo : thus with thre . values there will be'no difference higher than the second] i xq + 2h tae " i y does ahh. 291 - Yo) + (y2- 291 +90) h = 7g 0+ 401 +92) ee ty + 4h h’ - Similarly, xg +2n 7 dx = g 2+ 402 +94) 24g + nh h Speman & =3 On-2+ 4%n-1+ In) : (when nis even) On adding all these integrals, we get +nh h i ‘ y dx = F100 +In) +401 +924. +In-v +22 +94 + ek In-DI This formula is known as Simpson’s one-third rule. Note : In this case, we have neglected all differences above the second so y will be polynomial of second.degreesonly: ie, y = ax? + bx +c. Q. 11. Explain Simpson’s 3/8 rule. Solution.'Simpson’s three-eight rule (for point formula) : 3 Put n = 3 in general quadrature formula and neglect all differences above the third. In this case only, four functional values will exist and consequently there will be no differences higher than the third. Thus, we get . ph Pe 8 y dx = Af sp048Gx"> ~90)+300- 294 +30)4208— 390 +2y4-0)] > = se byo + 891 + 892 +Ysl SM ixgson BR Similarly, See an 2 = 3 lys'+ 3y1 + 8y3 + Yol he 3h sere an 1 aE = gy On-9 4 Bn-2# Dna + Ind Adding all these Been we have, when n isa multiple of 3. (225 ydx = ah [G0 +n) + 801 +92 +4 +5 ++ +¥n—1) ‘0 +2y3t+Ier- “+ 9n—a)] This formual is known as Simpson’s three- eight rule. ae in Q. 12, Evaluate I = Jo T sae geese is (i) Trapezodial rule, (ii) Simpson’s 1/3 rule and Gii) Simpson’s 3/8 rule. si »)° , on integrating by parts = (4-6) [log, (46) - 1] - 3flog, 3 - 1] =(4°6) [(1-52606) — 1] — 3[1:09861 - 1] = (46) (052606) - 3(0-09861) = 2-419876 — 0:29583 = 2:12405 whith Shows that the error in using Simpson’s rule is 0-00003. q. 4. &xplain Gauss’ elimination method. . Solution. This is one of the most widely used methods. This pethod i isa systematic process of eliminating unknown from the yinear equatidns. This method is divided into two parts : (a) Triangularization (b) Back substitution The system of n equations in n unknown is reduced to an quivalent triangular system (an equivalent system i isa system having jdentical solution) of equations of type 21% + QyQ%p + AygXg t+...+Gin%y = by AgoXq + Agg%y +...+ dan Xn = be ‘ (1) QggXg +...+Ogn%p = bg AnnXn = by This new equivalent system of equations can be easily solved by using the back substitution procedure: The details of the Guass elimination method are given below : Step 1. (i) Eliminate x, from 2nd equations onwards. By using the recursion formula R;- 28. Rego i = 2, 3,4, 0.0 THEY 3 (ii) Subtract from the second equation oat times the first AS » 1 ee equation i.e., ( ~21 RB, fori= 2) . This results in 1 4 : Qo; Gar [oa ~ "Aha X41 +| dog ~ Shag |e tt [ Ay |Xp Hy 7 c ‘41 ‘11 sm yy which on simplifying gives equation of type Gogg + AggXg + ++. + AanXn = bo. (ii) Subtract from the third equation 2 times the niet ‘uation, i ie,"

You might also like