cw R Numerical Differentiation and Integration 9.1 NUMERICAL DIFFERENTIATION Consider a set of values (x, y,) of a function. The process of computing the derivative or derivatives of that function at some values of x from the given set of values is called Numerical differenti This may be done by first approximating the function by a suitable interpolation formula and then differentiating it as many times as desired. ‘ If the values of x are equispaced and the derivative is required near the beginning of the table, we employ Gregory-Newton forward interpolation formula. If it is required near the end of the table, we use Gregory-Newton ‘backward interpolation formula, For the values near the middle of the table, the derivative is calculated by means of central difference interpolation formulae. If the values of x are not equispaced, we use, Newton's divided difference interpolation formula or Lagrange's interpolation formula to get the derivative value. 9.2 DERIVATIVES USING NEWTON'S FORWARD DIFFERENCE FORMULA Newton’s forward interpolation formula is 1 -IMp-2) oe D ay, + PE at yt 9.1) Y=, t PAyyt where p= nll Scanned with CamScannerNumerical Differentiation and Integration 9.7 example 9.1 Find the first, second and third derivatives of f (2) at xel5if x 1.5 2.0 25 300 3.54.0 f@) 3375 7.000 13.625 24,000 38.875 59.000 Solution Wehave to find the derivative at the point x= 1.5 which is at the begifining of the given data. Therefore, we use here the derivatives of Newton’s forward interpolation formula. The forward difference table is as follows: x y=f@) ay ay ay ay 15 3.375 3.625 2.0 7.000 3.000 6.625 0.750 25 - 13.625 3.750 10.375 0.750 3.0 24,000 4,500 14.875 0.750 35 38.875 5.250 20.125 40 59.000 Here, x,= 1.5, y, = 3.375, Ay, = 3.625, A¥y,=3, Aty, = 0.75 and h= 0.5 Now, from Eqn (9.4), we have 4 1 4 =f") = Gldn= ZAP LB Im GARE I i of! (15)= é [3.625 — 30) + 5 (0.75)]=4.75 ° From Eqn (9.6), we have ? 2 =f" @)= a (ay, - +E sti mee] 1 SF LS)= Gee B-0751=9 Scanned with CamScanner9.8 Numerical Methods Again from Eqn (9.7), we have 3 Hz Aye] : 1 ff" (1.5) = (5p (-75)=6 Example 9.2 The population of a certain town (as obtained from censug data) is shown in the following table. Year 1951 1961 1971 1981-199) Population (in thousands) 19.96 36.65 58.81 7721 94.61 eel Find the rate of growth of the population in the year 1981. Solution Here, we have to find the derivative at 1981 which is near the end of the table. Hence, we use the derivative of Newton’s backward difference formula. The table of differences is as follows: x y wy vy vy oy (year) (population) 1951 19.96 16.69 1961 36.65 3.47 22.16 -9.23 1971 58.81 3.76 11.99 18.40 2.76 1981 77.21 -1 17.40 1991 94.61 Hence, h= 10, x, Vty, = 11.99 n= 1991, Wy, = 17.4, V¥y, =—1, Wy, = 2.76 and We know from Eqn (9.10) that 2 ¥ 1 2p+l Ppa Bhp ty 6022 + sop? eae aa ae 0 Scanned with CamScanner‘Numerical Differentiation and integration 9.9 aa have to find out the rate of growth of the population in the = 1981-1991 _ 10 ae. =I, A= 10 and the values of Vy,, V¥y,, V'y, and V4, in #) | : ie Fe | roe ic.x, + ph= 1981 -. p 2(-1)+1 Sea 1 2 y"(981) = 7g Ll74+ + SEED? 276) : 2(-1)3 + 9(-1)? +11(-1) +3 12 = 1/10 [17.4 + 0.5 ~ 0.46 — 0.999166] = 1.6440833 “. The rate of growth of the population in the year 1981 is 1.6440833. (11.99)] Example 9.38 Obiain'the value of f'(90) using Stirling’s formula to the following data: x 6 75 90 105 120 ‘f@) 282 38.2 43.2 409 37.7 ‘Also find the maximum value of the function from the data. ‘Solution Since x=90 is in the middle of the table, we use central difference formula and in particular, Stirling’s formula. The central difference table is as given below. x vase) Ay ay ay Aty 60 282 10 75 38.2 5 5 23 90 43.2 “13 87 23 6.4 105 40.9 09 3.2 120. 37.7 Here, x,=90, yp =43.2, Ayy=23, AV. = 5, AV. h= 15. 23, A¥y.=6.4and 4 Scanned with CamScanner9.10 Numerical Methods Now, from Eqn (9.18), 4h He Ps} ron 539-444] = Wis [1.35 - 0.3416666] = 0.0672222. To find the maximum value of the tabular function: By Stirling’s formula, 2 2 B Yat PH =I+ 5 (Ay, + AY.) + Dr AYy, ae) [eet]. BRP yy Substituting the values from the table, we ee after simplification, 3.2 + 1.35 p—3.65 p? + 0.3417 (p -p) or y = 0.3417 p? — 3.65 p? + 1.0083 p+ 43.2 Ify is maximum, dy/dp = 0 i.e. 1.0251 p?- 7.3 p + 1.0083 =0 _ 73+ 03" —4(1.0251)(1.0083) ~ 2(1.0251) P= 6.9803 is out of range. . p= 0.1409 Hence, x= x, + ph=90 + 15 (0.1409) = 92,1135 and maximum of y = 0.3417 (0.1409) — 3.65 (0.1409)*+ 1.0083 (0.1409) + 43.2 = 43.27 = 6.9803 or 0.1409 Example 9.4 Using Bessel’s formula, find the derivative of f(x) atx=3.5 from the following table, x__347 348 3493.50 351 3:52 3.53 F@) 0.193 0.195 0,198 0.201 0.203 0.206 0.208 — Scanned with CamScanner9.16 Numerical Methods 9.6 NUMERICAL INTEGRATION (‘The process of computing {*y dx, where y = f(x) is given by a set of ibulated values [xp y,}. /= 0, 1, 2, «sm a= %, and b= x, is called numerical integration. Since y = F (®) is a single variable function, the process in general, is known as_quadrature. Like that of numerical differentiation, here also we replace f(x) by an interpolation formula and integrate it in between the given limits, In this way, we can derive formulae for approximate integration of a function defined by a set of numerical values. 9.7 GENERAL QUADRATURE FORMULA In this section, we will derive a general quadrature formula for equidistant ordinates. Let I= [Py de, where y={(x) takes the values yy Vy Vy. ++I, forty Xp Xp +++ Let us divide the interval (a,b) into n equal parts of width h, ‘so that = yx, =) +h x= 3, +h... %,= + Mh=b Y OX) -Xgsh Xp ah Then [= {2° f(x)de Putting x= x, + ph, so that dx = h dp, in above, we get T=h f) £0+ Ph) dp=h fy, dp a Scanned with CamScanner9.24 Numerical Methods 9.12.3 Errors in Simpson's 3/8 Rule and Weddle’s Rule Proceeding as above, the principal part of the error: s i) for Simpson’s 3/8 rule is a. Ly" in the interval [x,,x,] iW Gi) for Wedale's rule is~ 77> y" in the interval [3, x, Example 9.6 Evaluate > =F by using (i) Trapezoidal rule, i) Simpson’s 1/3 rule, i) Simpson’s 3/8 rule, and (iv) Weddle’s rule, Compare the results with the actual value. Solution Taking h= 1, divide the whole range of the integration [0, 10] into ten equal parts. The values of the integrand for each point of sub-division are given below: 1 Y= Tyg? 15% O5S=y, 0.2=y, O.1=y, 0.0588235 0,0384615] 1+ TM Ys x 6=x, T=x, 8=x, 9=3, 10=x, 1 = Tyg? 1027027 0.02 —0.0153846 0,0121951 9.9009901x10" “% =% () By Trapezoidal rule: 10 Lh f Ta 7 Oot Md) +2, FY, AY AY AY AY AY, ARTI 1 = ZI +9.9009901 x 109) +2005 +0.2+0.1 + 0.0588235 + 0.0384615 + 0.027027 + 0.02 + 0.0153846 + 0.0121951)] = 1.4768422 — Scanned with CamScannerNumerical Diterentiaon and Integration 9,25 (ii) By Simpson's 1/3 rule: lnv* él Flo,+70+46, *HtY+Y, AY) ive +20, +H, t Yt) 1 = 7 [Kl +9.9009901 x 109) +4005 +0.1 +0.0384615 + 0.02 + 0.0121951) +2(0.2 + 0.0588235 + 0.027027 + 0,0153846)] = 1.4316659 ai ae 3/8 rule: fae 2 HOt 40 +3 O FY, AYA tH tH) 420, + Yet Ye)] is [C1 + 9.9009901 x 10°) + 3(0.5 + 0.2 + 0.0588235 + 0.0384615 + 0.02 + 0.0153846) +2(0.1 + 0.027027 + 0.0121951)} = 14198828" (iv) By Weddle’s rule: is ae =z 200, FI) +, +I, +I.4 5+ 29+ 57, +y,+ 6y,) “8 = 2 [C1 + 9.9009901 x 10) + 5(0.5)+ 0.2 + 6(0.1) + 0.0588235 + 5(0.0384615) + 2(0.027027) + 5(0.02) + 0.0153846) + 6(0.0121951)] = 14410924 = tan x] = tan'10 = 14711277 Which shows that the value of the integral found by Weddle’s rule is the Nearest to the actual value than others. Scanned with CamScanner9.26 Numerical Methods Example 9.7 The velocity v of a particle at distance s from a point on its path is given by the following table : sft) 0 10 20 30 40 50 60 vifds] 47 S864 65 615238 Estimate the time taken to travel 60 ftusing Simpson’s 1/3 rule, ‘Compare the result with Simpson’s 3/8 rule. Solution We know that the rate of displacement is velocity, i.e. ds/dt= vy, ‘Therefore, the time taken to travel 60 ft is given by = Bape fee] where s =x and y= I/v, The table is as given below: xe) 0 10 20 30 Y= Vy 0.0212765. 0,0172413 0.015625 0.0153846 = = =r, % x9) 40 50 60 yal 0,0163934 0.0192307 0.0263157 7 =y, =% By Simpson's 1/3 rule : h pa = G10 *H) 420,49) +4(y,4 y,4y,)] 10 = sl (0.0212765 + 0.0263157) + 2(0,015625 + 0.0163934) +4(0.0172413 + 0.0153846 + 0.01 '92307)} = 1.063518 ' Hence, the time taken to travel 60 fis 1.064s, By Simpson's 3/8 rule : oy 3h dx. 3h p = 3 lOoty) +30, ++, + Ys)4+2y,) 30 = “g [(0.0212765 + 0.0263157) + (0.0172413 + 0.015625 + 0.0163934 + 0.0192307) + 2(0.0153846)] = 1.0643723 By this method also the time taken to travel 60 ft is 1.064s. oo ac Scanned with CamScanner0 NUMERICAL DIFFERENTIATION Numerical differentiation is the process of calculating the derivatives of a given function by means of a table of given values of that function. That is if (xj, y;) are the given set of values, then the process of computing the values of 2 4 y+ efe. at a given point is called numerical differentiation. If the values of the derivatives are required at a point near the beginning or the end of the table, we use Newton-Gregory forward or backward formula, If the values are not equally” spaced first find the polynomial by using Lagranges formula and then find the derivatives at a particular point. OQ FORWARD DIFFERENCE FORMULA TO COMPUTE THE DERIVATIVES - Newton’s forward interpolation formula is r(r-1) r(r-1) (7-2) FO+trh) =yotrAyot gy A? y0+ 31 3 yo r(r=) (-2) (7-3. + A At ygt ee (Here using of r for n is only for convenience) Differentiating w.r.t. ‘r’, we get, = ene dy A? yp aoe? By = : 2 9+ Ur=3 Atygt... (1) WF "Gq+rh) = AP yq+ (r-1) AY Part yyy. + @ Faget) = Bypt 2S ahyp+ ~@) ea Pr te Scanned with CamScannerNUMERICAL METHODg 32 ; Similarly we can find the remaining derivatives. If we want to find the derivatives at a point x = Xo, then Xt rh=xo ie, r=0. i Hence on substituting this value of r = 0 in the above formula (1), (2) and (3), we get 1 1 t f’@) = §[40-742%0+5 490-4 At yg | 1 u F(a) = Bl 42-48 yo +77 4* y0- | FG) = ala Yo -3 At y+ | and so on. Q BACKWARD DIFFERENCE FORMULA TO COMPUTE THE DERIVATIVES Newton’s backward difference formula is Oe Dye r(r+1)(r+2), F (my trh) = yotr Vyo+ ot 31 V3 yo A jes beaded) PEDO CHD ayy ay (Here using of r for n is only for convenience) Differentiating (1) w.r.t.‘r’, we get, 2r+1 32+ 6r+2 2 V4yq+ 6 v3 fot rh) = Vyg+ % AAs or + res v4 W £" (xq +rh) = Vey t (r+ 1) Viyp 4 9418+ HI vay, 4. 1B f"(xgtrh) = Vy +3 V4 ye. Similarly we can ‘nd (ne remaining derivatives. BUNIT3.., Scanned with CamScanner\ r [NUMERICAL DIFFERENTIATION AND INTEGRATION 3.3. : at. At the point x =xpriie., xgtrh = x9, we have r= 0. voy ed 1 1 1 £'G0) =h [¥0+3 V0 +5 V9 +E V9... ] a” al i Fa) = 7[V0+ V0 +B V0 +.-.] : 1 3 £0) = B[V +5 V4yy+ | and so on. STIRLING’S FORMULA TO FIND THE DERIVATIVES AT THE MID POINT (WITHOUT PROOF) : 1[{(Avot dy) 4 yo) -+(( 2 )~ 73 P41 +4%2) 1 -39 v2 + ASy_3 ti) 1 1 y"@o) = ar [2% -77 AY + Find the first, second and third derivatives of the function tabulated below at the point x = 1.5. [x | 15 | 20 | 25 | 30 | 35 | 40 7 @) [3375 | 7.0 |13.625| 24.0 [38.875| 59.0 ‘The difference table is as follows : _ x ly=f@ Ay Ay A3y Aty | 1.5x9|3.375 Yo > 3.625 (Ay) 2.0 |7.0 3.0 (A?y9) 6.625 0.75 (Ay) 2.5 |13.625 3.75 = 10.375 0.75 3.0 |24.0 4.50 9 14.875 0.75 3.5 [38.875 5.25 20.125 40 |59.0 . UNIT3 & ere Se Scanned with CamScannera | NUMERICAL METHODS 34 ivative at the point x = 1.5 which j we have to find the derivative at . is the inital value of the table. Therefore by Newton's forward difference formula for derivatives at x = xg, we have 1 1 1 $%la9) = i av0-7 279043 Ady - ral Here xp = 1.5, 8 =0.5 * FS) = a5 [os -46.0)+4.00.15)| 05 2 3 f'U.5) = 4.75 Atthepoint x=xo, 1 ia FG) = B[a%- a+b aby Here xp = 1.5,h=0.5 2 fU5) = os (3.0-0.75) “A1.5) = 9.0 Atthe point x= Xy (eo “Hm Fe (1.5) = _1 07) (sy = 60 rr BUNIT3..,, — > | Scanned with CamScanner| \ [NUMERICAL DIFFERENTIATION AND INTEGRATION __ 3.6. ! EM Example 2 RY F 2, From the following table, obtain the value os at the point x= 0.96. A R \ x | 0.96 | 0.98 | 1.00 | 1.02 | 1.04 |, () | 0.7825 | 0.7739 | 0.7651 | 0.7563 | 0.7473 ‘The difference table is as follows : g y Ay dy By Aty 0.96 (x9) [0.7825 0) -0.0086(Ay9) 098 [0.7739 -0.0002(47y9) |-0.0088 10,0002 (A7y9) 1.00 [0.7651 0 0.0004 (A*y9) /-0.0088 0.0002 102 |0.7563 0.0002 |-0.0090 1.04 |0.7473 Using Newton’s forward difference formula for derivatives at the point x = x9, we have 1 U y”@) = a 40-450 12 40- a eee Here xp = 0.96, h = 0.02 — 1 i £90) = Toge &0.0002 = 0.0002 + 73 (- 0.0004) ] 0.0002 — 0.0002 - 0.000366] sual 3 = goood & .. UNIT3 & Scanned with CamScannerNUMERICAL METHODs 4 Tb d second derivatives of the function ind the first an : 6 i ek a i a atx =3.0 using the values given below. x 3.6 3.0 | 32 34 3.6 38 | 40 f(s) _|=14 — 70.03 | —5.296 | 0.256 _| 6.672 14 The difference table is as follows : x f@) Ay | Ay | Ay Aty | A’y 30 | —14 @o) Oo) | 3.97 (Ayo) |-0.764 ' 3.2 | -10.03 (Ayo) | 0.054 ' 4.734 (A3yp) | - 0.008 34 | -5.296 0.818 (A4Y9) |.0.006 |. 5.552 0.046 (A%y0) |" ' 3.6 | 0.256 0.864 — 0.002 6.416 0.044 3.8 | 6.672 0.908 | : 7.328 ' 4.0 4 Here we have to find the derivative at the point x = 3.0 which is the initial value of the table. Therefore by Newton’s forward difference formula for derivatives at x = xq , we have ’ 1 f' (3.0) = 02 [397-4.0.768) +4 0.054) 1 4 (- 0.008) +5 0.0) 5 (3.97 — 0.382 + 0.018 + 0.002 + 0.0012) = 53.6092) f' 3.0) = 18.046 f' B.0) Scanned with CamScannerNUMERICAL DIFFERENTIATION AND INTEGRATION 3.7 ee fF At the point x = 9 £00) = 98 [8% 0— A v0 +L aby | Here xp = 3.0, h = i " =e £80) = Gp [ 0.764 - 0.054 + 4 (- 0.008) ] = 25 [0.71 - 0.00733] = 25 (0.7027) ; Obtain the value of f (0.04) using an appropriate formula for the einer data: 0.01 0.02 0.03 0.04 0.05 0.06 Fen 0.1023 | 0.1047 | 0.1071 | 0.1096 | 0.1122 | 0.1148 The difference table is as given below. x _|y=f@)| Ay Ary Ay Aty Ady | 0.01 | 0.1023 (zo) | Gv) | 0.0024 0.02 | 0.1047 (Ayo)}| 0.0000 (0.0024 | (4%yq)|0.0001 0.03 | 0.1071 0.0001 (Ayq)|— 0.0001 0.0025 (A*y0)|0.0000 0.04 | 0.1096 0.0001 |0.0000 | -0.0001 (4°y0)} 10.0026 0.05| 0.1122 0.0000 |-0.0001 10.0026 0.06 | 0.1148 Here xp =0.01 and h=0.01 Newton’s forward difference fom for first derivative is 72 2 L Ary 4 3 = Gre? 43 naanaleiys + 12 Y(t rh) = 1 ayo +5 Atyg+ sa} . UNITS & a liieeaeineieaiearcrmeeae adnan ennatetatemeeetatte tte Scanned with CamScannerBeer ay NUMERICAL METHOns y (0.04) = ? Xp trh = 0.04 0.01 + (0.01) = 0.04 r=3 ! 6 y"(0.04) = ar (0.0024 +52 (0.0000) +0) 1 = aor (0.0024) (0.04) = 0.24 PM Example 5| For the given data : x | 10 | 14] 12 | 1af 14] 15 | 16 Feo: | 7.989 | 8.403 | 8.781 | 9.429 | 9.451 | 9.750 | 10.031 Fina and “Ga atx= LL. The difference table is as follows : x |y=fa)| ay | a% | ay | ay | ay | ay AI Go} 7989 . (0.414 1.1} 8.403 | (Ayo) |- 0.036 0.378| (Ay9) | 0.006 1.2| 8.781 cmp | Soup ral oz |°8} ~ 0930 oon | 290 [00 | 0.322} - 0.026 = 0.001 | (4°) Sy) 14| 9.451 ioe 0.003 |¢ 0.299] — 0.023 0.002 15] 9.750 ane 0.281| -o.o18| | 0 1.6} 10.031 Here we have to find the derivaii i ich is tere lerivative at the point x = 1.1 which i the initial value of the table. Therefore by Newton’s forward derivatives at x = %qr We have difference formula for mUNIT3.., Scanned with CamScannerNUMERICAL DIFFERENTIATION AND INTEGRATION 39 1 1 f'@o) = tLar—Fa% +405 =| Herexp=10,h=0.1 a 1 2 fl) = 7 (0414-3 0.036) +4 0.006) Using Newton’s forward difference formula for derivatives at the point x =o, we have om 1 “9 GO)= Toil- 0.036 - 0.006 + 15 (~0.002) +.) = - 4.3833 mn Ay »"Go) = 52 [470 ~ A3yg + 4 Atyy— Here x9 = 1.0, h = 0.1 — 4.3833 The following data gives the velocity of a particle for 20 seconds at an interval of 5 seconds. Find the initial acceleration using. the entire data, Time (sec) : 0 5 10 15 20 Velocity (m/sec) : |_0 3 14 69 | 228 The difference table is as given below. t=(x)|v=(y)| Ay Ay Ay Aty 9 ()| 9 Oo) 3 (Ayo) 5 | 3 8(A?y0) ‘ ll 36 (A*y0) 10 |14 44 24(A*yo) 55 60 15 |69 104 159 20 _|228 seve UNITS Scanned with CamScannerage MEMO a REHM Here x= 0; h=5 Since wé want to find acceleration when ' = 0 which is nearer to the beginning of the table, we use Newton’s forward difference formula for first derivative. We know that Newton’s forward difference formula for first derivative is 1 1 1 i #700) =} [490-3 429043 4 90-440 i Tf ‘V’ (=y ) is velocity, then x =y (Xp) ] is acceleratior.. Now we have to find the acceleration at = i.e, x=0 (G7), 77 OaF 260 = a3 -4@) +466) = Jem | t (3-44 12-6] yO =1 :. Acceleration when t = 0 is 1 m/sec’, 4 a es Find the value of sec 31 from the following data. 0 (deg): 31 32, 33 34 tan@ : | 0.6008 0.6249 0.6494 0.6745 The difference table is given below. x y Ay Ary Aty 31 (x) | 0.6008 (yo) 0.0241 (Ayo) 32 | 0.6249 10.0004(Ay9) 0.0245 jo.o00% 33 | 06404 0.0006 (a0) 0.0251 34 | 0.6745 BUNIT3...... 4 Scanned with CamScannerFee Here #= 1, x0=31 and y’(tan 6) = sec? @ Newton’s -forward difference formula to find derivatives at x =p is given by p a J Vay cag f'@o) = | A¥0- 24° 0 +5 A v0 GAty9...] 1 1 y’Gl) =se0 31 = Fo" [0.0241 ~7 (0.0004) +3 (0.00029 180 ae [0.0241 — 0.0002 + 0.000066] sec? 31 = 1.36936 sec 31_= 1.1702 @ Eee » The table given below reveals the velocity v of a body during the time ‘t? specified. Find its acceleration at t = 1.1 x | 10 11 12 13 14 y_| 43.1 | 47.7 | 52.1 | 56.4 | 60.8 The difference table is as given below. [t=@) [v= | ay [ae Ady ty 1.0 (%q) [43-1 Oo) 4.6 (Ay) Li (47.7 _ | 0.2 (Ayo) 1 44 0.1 (Ay) 52.1 5 43 a 0.2 0.16490) 13/564 0.1 44 | 14 (608 Here xp = 1.0; h=0.1 Since we want to find acogleration when t = 1.1 which is nearer to the beginning of the table, we use Newton’s forward difference formula for derivatives. . UNITS Scanned with CamScannerNUMERICAL METH oe We know that Newton’s forward difference formula for first derivative is 2r 1 =1 49, , 3 =6r+2 y'p+ 7h) =% Ayo + A’yo+ 6 4% 23 - 972 + Ir-3 ae 12 = aby wv \ If ‘V’ (= y ) is velocity, then Gp [ = y'@o + rh) Lis acceleration. Now we have to find the acceleration at ¢= 1.1 i.e. x= 1.1 yl) =? ie, xgt+rh = => = =d[ sordcon—teansdon| =O | 46+3(0.2)-G ON) +7900. = 10[4.6-0.1 - 0.0166 + 0.0083} = 44.917 «. Acceleration when t = 1.1 is 44.917. 4 Tee & ee oe knee oe fo r which f(x) is maximum in the range of iven, using the followir . panei value of f (x). Following table.. Find also the *: 9 10 1 12 | 13 | 14 y | 1330 | 1340 1320 | 1250 | 1120 | 930 The table of difference is as follows : SUNT... Scanned with CamScannerre \ NUMERICAL DIFFERENTIATION AND INTEGRATION 3.43 oa——__. eS —— x |y=f@ Ay A2y a 7 A 9 {1330 y 10 | 10 (xo) |1340 Oo) -30 | — 20 (Ayo) -20 | {1320 = 50 (A2yo) 10 | -70 : -10 \ 12 {1250 60 (A3y9) | 10 - 130 13. {1120 -60 0 - 190 14__|930 By Newton’s forward difference formula for derivative we get yf (ig #rh) = Hayy +255 oe aye SA Or42 43 Gr+2 2 42be 9% joe 3 aly, a i 0) ce where r= Atthe maximum point of x, y' (xg +rh) = 0 Here x= 10; Ayg=—20; A’yg=-50, A” yg =- 10and h= 1. ‘Therefore equation (1), becomes 429250) 3 rt GC! 0 | ¥ G+ rh) = vps a SAY, 5) [eso] 3 = -20-50r+25+ VOq+rh) = Maximum value of x is obtained by the equation y'(xq + rh) = 0. t From equation (2), we get =ish— 120045 tien) | UNITS | | i Scanned with CamScannerNUMERI 34 CAL METHO0 152 - 120r+5 3 -157-120r+5 537 +24r-1) 3P4+24r-1 = Solving equation (3), we get 24 VOD AGM _ -24 +5764 f= 23) 7 6 r =-8.04; 0.0415 ooo °o +) We know that, x-%9 Toe tt Substituting r= — 8.04, x9 = 10 and h = 1 we get, x= 1.96 which is not acceptable, where as’ when we are substituting r = 0.0415, xo = 10 and h =I we get, x = 10.0415 which is acceptable. Therefore y = f(x) attains its maximum value when x= 10.0415. Next we have to find the maximum value of y when x= 10.0415 Newton's forward formula is yiq+ nh) = yytnAyy+2A—D oT D ny n(n = 10-2) + 31 A*y9 Picen y(10.0415) =? ie, xq +nh= 10.0415 Here x = 10 and h = 1, we get 10+n= 10.0415 0.0415 - 10 Jon = 0.0415 910.0415) = y9 + (0.0415) ayy + (0.0415: oust AY 4 (0.0415) 0.0415=1 0.0415 =2 iygt en @ UNITS Scanned with CamScannerMUMERGAL DIFFERENTIATION AND INTEGRATION 3.5 y(10.0415) = 1340 + (0.0415) (-20) + 0.0415) 0.9585 50) (0.0415)(~0.9585)(-1.9: ‘y(10.0415) = 1340.03 Hence the maximum value of f(x) is 1340.03. 4 é Find the values of sin 18° and sin 45° from the following table. x° 0 10 20 30 40 =cosx° | 1.0000 | 0.9848 | 0.9397 | 0.8660 | 0.7660 Pa ere ‘The difference table is as given below. x | y=f® Ay Ay Ay Aty 0 (ap) {1.0000 (yo) } 0.0152 (Ayo) | - 0.0299 Y 10 [0.9848 (A2yq) {0.0013 ‘ — 0.0451 (A%y9) | — 0.0286 O.no10 j 29 0.9397 10.0023 | (4*y) - 0.0737 |-0.0263, j 30 |0.8660 0.1000 l 40 |0.7660 Here xy =0 and h= 10° Newton’s forward difference formula for first derivative is - 3r2—6r+2 Y (y+ rh) = tayo « 255 SEE? 3, * 2-97? + Mr-3 Atyy + ] te 4 Scanned with CamScanner46 NUMERICAL METHODg BAB pea NUMERICAL METHODS, y (18°) = ? . is, xytrh= 18 => r=49=18 4 ys) = al = 0.0152 + 1.3 (-0.0299) + ¢ (0.92)(0.0013) +c 1.392)(0.0010 | = 18(~0.05393) = -0.3089 Since y =cos x ; Be sinx c.=sin 18° = — 0.3089 sin 18° = 0.3089 y45°) = 7 myths se = r= 4 245 Ble 0.0152 + 4 (0.0299) +3 (35.75)(0.0013) de ] +34 (93)(0.0010) - 8 ude = 7 (- 0.12318) ce 107 mn! — 0.7058 sin 45° = 0.7058 EM Example 11 FY ' ie Population of a certain town is shown in the followin8 table. | Year 1931 | 1941 | 1951 | 1961 | 197! Population | 40.6 | 60.8 | 79.9 | 103.6 | 1327 (in thousands) Find the rate of growth of the ‘population in the year 1961. The table of difference is as follows : @UNITS.... Scanned with CamScannerAUMERICALDIFERENTATIONANDIWTEGRATION 47 eat) Popatain) Y w vey wm. 1931 |40.6 20.2 1941 |60.8 1.1 19.1 5.7 951 |79.9 4.6 I-4.9 23.7 10.8 V3: 4, 1961 |103.6 5.4 V2yp Yaa | 7 29.1 Vyo 1971 x9 [132.7 ¥0 Here h = 10, xo = 1971. Here we have to find the rate of change of the population in the year 1961 which is at the end of the table. Hence we lave to use Newton’s backward formula for first derivative We know that 7 1 2r+] 32+ 6r+2 y¥ @o+rh) =t[ v0.47 V2yq 4 5 ** Viy9 25 +9r? + LIr+3 {PEA EUS v4] W) We have to find the rate of change of population in the year 1961. ie, y' (1961) =? 1961 ie, xgtrh 1971+r-10 = 1961 ; . ; 1961-1971 _ ie, rms 19 =- Substituting r = -1, h = 10 and the values of Vyo, V2vq. V3¥0 and V4yg (from the table) in (1), we get y (1961) = 4 [29.1 3) 6.4) 2) (0.8) + @) 49] = 4 [29.1-2.7-0. 1333 + 0.4083] = 2.6675 <. The rate of growth of the population in the 2.6675 thousands year 1961 is .» UNITS Scanned with CamScanner3.18 NUMERICAL NETHong SA. = ‘A rod is rotating ina plane. The following table Bives angle 6 (radians) through which the rod has turned for at values of the time ‘t’ seconds, us t 0 0.2 0.4 0.6 0.8 8 0 0.12 | 0.49 | 1.12 | 2.02 3.20 Calculate the angular velocity and the angular acceleration, of the rod when t= 0.6 seconds, The table of difference is as follows : t(=x)| O(=y)|_Vy V2y v3y | vy | 0 0 0.12 02 | 0.12 0.25 0.37 04 | 0.49 0.1 0.26 06 | 112 | 98 01 0 0.27 0.90 10 08 | 202 3 02g |O1V3y0) 1.18 (V2yo) 1.0 x9 | 3.20 yy} (V0) Using Newton's ba i for first derivatives, we hong ckward difference formula Here h= T a! xa lh ; 06. nd the angular velocity when ¢ = 0.6 seconds, ie» ® is y'(0.6) = 2 Xotrh= 06 » 104702) = O6 WUNIT3._ | Scanned with CamScanner—_ \ NUMERICAL DIFFERENTIATION AND INTEGRATION 3.49 ie, Y ==2 ; 2(-2)+0 y 0.6) = Gas) (0.28) ¢ @ (22 + $ (-2)+ 2) ©. »] 1 = 02 {1.18 - 0.42 + 0.033] y’ (0.6) = 3.966 radians / sec Hence the angular velocity is 3.966 radians / sec To find angular acceleration : We know that ww Teo 1 yor rh) =e [V2y9 + (r+ 1) V3y9 + ---] 1 y”(0.6) =O27 [0.28 + (2+ 1) @.1)] 1 = 004 [0.18] y”0.6) = 45 radians / sec? Hence the angular acceleration is 4.5 radians / sec”. 6 EOE ¢ Consider the following table of data: x 0.2 0.4 0.6 0.8 1.0 Se): | 0.9798622 | 0.917710 (0.8080348 | 0.6386093 0.3483735 Find (0.25) using Newton’s forward difference approximation f (0.6) using Stirling’s approximation and £10.95) using Newton’s backward difference approximation. ee Newton’s forward difference approximation = The difference table is as follows . UNIT 3 4 Scanned with CamScanner| ae | 320 NUMERICAL METHOps 20 ______AMERICAL ME THONG x | y=f@) ay dey A3y [at] 02 | 09798622 0) |= 0.0620912 7 (yo) | - 0.047645 04 | ogi777io |-0-1097362| (4x9) | - 00120443 = 0,0596893 | (A3yq) | -0.049076] 0.6 | 0.8080348 | ~ 0-1694255 (yy ~ 0.1208103 | _ 9.961121 0.8 | 0.6386093. | ~ 92902358 0,3483735, Here x»=0.2, yo=0.9798622, h=0.2 The Newton's forward difference formula is 6 1 2r-1 3r2-6r+2 ; y’Gig+ rh) = 5 [Avy + FZ Ary + Go aby + | y’.25) = 2 ' ie, 4g trh = 0.25 ' ie, 0.2+n (0.2) = 0.25 ' ie, p= 235-02 _ O95 02 y/(0.25) = a [-0.0620912 + 21025)—1 (-0.047645) 4, 310.25)" - $025) +2 — 9.912043) J 1 = 93! ~ 0.0620912 + (-0.25)(- 0.047645) + (0.11458)(- 0,00138003) ] 1 = 07 0.051559] 9 (0.25) = - 0.257799 MUNIT3.... | Scanned with CamScannerNUMERICAL DIFFERENTIATION AND INTEGRATION 3.21 Gi) Stirling’s approximation : Y=f (x) Ay 2 x Ary sy Ady 02 | 0.979862 - 0.0620912 04 | 09177710 ~ 0.047645 ~ 01097362) (42) 4) | -o.0120443 .6 | 0.8080348 0 z ~0,0596893 | (A3y-9) | ~0.0890767 G0) -0.1694255 (a4y 9) (02y4) | -0.061121 (yo) a r — 0.1208103 (A-y_4) 98 | 46386093 | - o.2902358 : (A2y9) 10 03483735 Here x = 0.6, yo = 0.8080348 and h = 0.2 Stirling’s difference formula is 1[(Avot Avr) 1 y’Q%) = (Poses ~ Jz (91 + A¥yp) rene +A’y54...)] 39 (A’y2t dys +. 1_[/-0.1694255 - 0.1097362 y6) = 35| ee) 1 -pe 93 [ -0.13958085 + 0.00609 } y"(0.6) — 0,6674187 0.061121 - 0.0120443) ] . UNITS Scanned with CamScanner3.22 ah NUMERICAL METHoDg (iii) Newton’s backward difference approximation : x | Y=f@) vw vy vy | 02 | 0.9798622 i — 0.0620912 04 | 09177710 - 0.047645 -0.1097362 | -0.0120443 0.6 | 0.8080348 - 0.0596893 = 0.1694255 0.061121 t . 0.8 | 0.638693 -0:1208103 } (Y3yq) id =0.2902358 2, ‘ (V-yq) + 10 | 03483735 (Wyo) | &o 60) Here x9 = 1.0, yo = 0.3483735 and h = 0.2 Newton’s backward difference formula is y’Gy+r) =| [Vy9+ 2r +1 3 Viy5 422 ry Wy _ 2i+9% + lines vtyp + od Xo+rh = 0.95 y/(0.95) =? Xo + rh =0.95 1.0 +7(0.2) = 0.95 => rf to eae 1 ra 90.95) = 93 [-0.2902358 425 925)*1 9.120810) +3028) + 66.025) 42 constant] 1 = 9.9 (-0.2902358 — 0,030202575 0.007003) y'(0.95) = - 1.63764 @ UNITS... _ atl Scanned with CamScannerNUMERICAL DIFFERENTIATION AND INTEGRATION. 3.23 Se ae ee & é A function is according to the table given below. _ x 0.35 | 0.40 | 0.45 | 0.50 | 0.55 | 0.60 | 0.65 Fc) [4.521 | 1.506 | 1.488 | 1.467 | 1.444 | 1.418 1.389 | ‘the value of f ’(0.5). The difference table is as given below. - x |y=fe) | ay | ay |. ay | ay | ay (0.35(%_3) 1.5210_3) oe eee -o.015 0.40 (xg) |1.5060-9) [0.003 2)]15080-2) | oorg lo |-0.003 lo.001 (045 (1) |1-4880-1) | goat ‘0.001 |-0.003, a) (03y_2) (a>y_3) Looo | “7 |-oooe |" 7 i 10.50 (x) |1.467 9) Ea Fe | |-0.023, , ‘ayy o0ns | AD Joos |(4%2) 0.55 (x1) |1.444 0) 0.026 ~ 0.60 (x2) |1.418 (9) iene |-0.029 10.65 (x3) |1.389 (y3) ‘Here we have to find the derivative at the point x = 0.5 which is in the middle of the table. Therefore we apply Stirling’s formula for derivatives. Here h = 0.05, and take xp = 0.50 We know that the Stirling’s formula for the first derivative is voy (Avatar) 1 1 ‘| | SOs 1{Porou 5 La (a8y_1 + Aya) + 39° 24A5Y.2) | Lf 1 = —_|~(-0.023-0.021)-— (-0.001+04 aia 300%3 0.021)-(-0.001+0.001) 1 as +30 (0.003 - 0.003) | UNIT 3 Scanned with CamScannereS i... | 3.24 NUMERICAL METHODS, 4 = Toy 00224040] = -0.44 2 $70.5) = -0.44 & s The, tion above a datum line of seven points of a road, ‘are is jelow. x 0 300 | 600 | 900 | 1200 | 1500 | i809 135} 157 183 201 205 | 193 y p14Q Find. = ‘gfadient ofthe road at the middle point, a ee The difference table is as given below. x | vo | ay/lay] 3r | ay | Ay 0 135 14 1300 149 6, 8 24 600 1570.) 18 -50 i (WN oe) -26 (a3y_) 70 (&5y_3) : ~ |!8(A99) 3, 5, 120 orgy |e? BH SD fo 160%. 100 a -16 205 rt J~12 WW ie 193 We know that gradient means 2 Here we have to find the gradient at x = 900, i.e., 2 atx = 900. Here it = 300. Take x9 = 900, yp = 183 Using Stirling’s formula for the first derivative, we get ‘AyorA: ¥"(x%) -il( a1 = 309 | (C8E2 1 1 =300L > 26-26) +35¢16+70)] 1 ~ THO.) + A3y.9) + Jas gaa] UNIT 3, a Scanned with CamScannerNUMERICAL DIFFERENTIATION AND INTEGRATION 3.25 300 [22 + 2.666 + 1.8] = 0.08822 [The gradient of the road at the middle point is 0.08822. CA Example 16 FY Find the first and second derivative of the function tabulated below at x = 0.6 x 0. 0.5 0.6. 0.7 0.8 Ly | 1.5836 _| 1.7974 | 2.0442 | 2.3275 | 2.6511 The difference table is as given below. x | y=f@) Ay Ay A3y Aty 0.4 |1.5836(9-p) 0.2138 0.5 1.797404) 0.0330 0.2468 (Ay_;) (0.0035 0.6 0.0365 |(A*y_g) | 0.0003 2.0442 (0) 2 G0) *" \0.2833¢Ayq) (4°v-1) [0.0038 | (4°v-2) 0.7 |.3275 Qi) 0.0803: /A*y_») ‘10.3236 (A2y) 0.8 |2.65110y2) Here h=0.1,%=0.6 and yo = 2.0442 Using _Stirling’s formula for the first derivative, we get [(AvorAy1) 1 y&o) = 5% (et } 7g ( Aty 1 +45y_2) 1 + Fahy gta'y ay..] a [(02ss2 02868) (0.0038 + 0.0035)] 10 [0.26505 - 0.0006083] 2.64442, so UNITS Scanned with CamScannerNUMERICAL METHODS y"(0.6) = oF [ a.036s -40.003)] = bil 00365— (0.0833) 0.008) = 100 {0.036475 } YO.) = 3.6475 ; y'@.6) = 2.64442 y"O.6) = 3.6475 & 6 Given the following data, find y’ (6). x fol[2|]3]4 [7/8 y: | 4 | 26 | 58 | 112 | 466 | 668 = |y=F@] agen Bra | Br@ | Mse 04 126-4 20 = | gy 2 126 3- H-7 58-26 _ 4 4-0 =} 3-2 54-32, 1-1 9 3 |s8 4-2 io-11_, [7-9 7-3 =54 | 4-3 = |g _ 54 1-119 4 ju 7 2 7 |466 668 ~ 466 8 |668 827 = 202 hb Newton's divided difference formula is Y= FO) = yy (xq) AF (xq) + (2-H) wa? sa + (=29) («= x1) @— x) BF G0 _ v Scanned with CamScanner= ~~ NUMERICAL. DIFFERENTIATION AND INTEGRATION ‘zr Here x9 = 9 Y0= 41 = 2.4953, Af (mp) = 1b 47 f Go =7, 43 FQ) =1 Substituting these values in (1) we get, fx) =44x4x(%-0) 11 +x(x—- 0) (x-2)(7) + x(a 0) (x-2)@—3) (1). = 44 1x + (XP - 2x)7 +x (2 - 5x + 61) = xt 7x? 5x4 Ix + 6x— 14x +4 y@) = +2 +3x+4 y/(6) = (6) +26) +3) +4 = 108+244+3 yO = 135 a= dae) L Cateutate ana at x = 1.35 from the following data. 5 re 1.2 13 14 15 16 y_] 1.62628 | 0.15584 | 2.45256 | 5.39168 | 9.12500 | 13.83072 [Ans. 29.34, 71.33] 2. A rod is rotating in a plane about one of its ends. If the following table gives the angle @ radians through which the rod has turned for different values of time t seconds, find it angular velocity and angular acceleration when t = 0.7 seconds. t seconds 0.0 | 02 | 04 | 06 | 08 | 10 8 2 fadians 0.0 | 0.12 | 0.48 | 1.10 | 2.0 | 3.20 [Ans. 4.5 radians, 7.25 radians/sec?] 3. The population of a certain town is shown in the following table. Year 7921 | 1931 | 1941 | 1951 | 1961 Population 1 (in 19.96 | 38.65 | 58.81 | 77.21 | 94.61 thousands) Find the rate of growth of population in 1951. Ans. 1. 79] vee UNITS ee oe Scanned with CamScannerNUMERICAL METHODS, 3.30, 0 NUMERICAL INTEGRATION f The term Numerical integration is the numerical evaluation of a definite integral 5 A= J fade a where ‘a’ and ‘b’ are given constants and f (x) is a function given analytically by a formula or empirically by a table of values, Geometrically, A is the area under the curve of f (x) between the ordinates x =a and x = b. But in engineering problems we frequently come across the integrals whosc integrand is an empirical furiction given by a table. In these cases we may use a numerical method for approximate integration. When we apply numerical integration to a function of a single variable, the process is sometimes called mechanical quadrature; when we apply numerical integration to the compttation of a doub‘e integral involving a function of two independent variables it is called mechanical cubature. The problem of numerical integration, like that of numerical differentiation, is solved by representing the integrand by an interpolation formula and then integrating this formula between the given limits. Thus, to find the value of the definite integral 5 5 J £6) de (09 | y dx we replace the function f(x) (or y) by a interpolation formula, usually one involving differences, and then integrate this formula between the limits a and b. In this way We can derive quadrature formulae for the aj i i ion ‘ n ipproximate integratiol any function for which numerical values are known. Of the many possible quadrature formulae, here we shall derive ti of the simplest and most useful one. QUADRATURE FORMULA - FOR EQUIDISTAN ORDINATES - Consider the Newton's forward difference formula iE . -1 YM=y Co +h) = yp + ndyy +2 A—D yp n(n—1) (= 2) A3yp 4 + 31 APyo* BUNIT3 Scanned with CamScanner3.34 NUMERICAL METHOng h By esi? heaiory AL = 4a, +4, FQ toe a , 7B ” = hy tor +7 yg i" + + (5) Q)-6)=> 2 1 1 ie yar a, = 3yq) Byte. a1 -h a Bvt. i.e., Principal part of the error in (x1, %9) 3 "at Similarly principal part of the error in the interval (x,.x3) =H = 77 Y2”and so on, 3 Hence the total error E = + Di" +924... +yn"] =nhé “—E< ——y” Dp” ® ‘Where y (6) is the largest of the n quantities Ys 2% ven” ; a= (b— a) h2 he, E< “Ty y"@ = ear y"@) bons +. Error in the trapezoidal rule is of the order h2. % el 52 z Compute the value of the definite integral} 1Ge} dx or 5.2 vee i J In xdx using trapezoidal rule. 4 Divide the interval width 0.2 ie, h = calculated for each : f Of integration into_six equal parts each o! 2. The values of the faction y =n x are nex! Point of subdivision as given below. ‘ WUNITS.., : Scanned with CamScanneryE NUMERICAL DIFFERENTIATION AND INTEGRATION 42 44 46 48 5.0 5.2 1.435084 | 1.481604 | 1.526056 | 1.568616 | 1.609437 | 1.648658 YL y2. 33 v4 ys 46 By Trapezoidal rule, we have 52 h ; J inxde= 9 (O0+Y6) +2 Or+y2 +93 +4 + 5) 4 0.2, = “q 1.386294 + 1.648658) + 2 (1.435084 + 1.481604 + 1.526056 + 1.568616 + 1.609437))] = (0.1) [3.034952 + 15.241562] 18276544 ay ata equal parts using trapezoidal rule. [Nov. ’91, Nov. ’89] Here the length of the interval is h = if = 0.25. The values of the function y = e~*” for each point ay are given below. * [0] 025 [05 | 075 | 1 2 | 1 | 0.9394 | 0.7788 | 0.5698 | 0.3678 By) v1 y2 33 Yq By Tapia rule we have 4 § Evaluate { e~ dx by dividing the range of integration i | 0 | | i ear = 4 (op+y)+2014499)] = 225 [1.3678 +2 2.2876) = (0.125) (5.943) 1 } [ede = 0.7428 UNITS & Scanned with CamScanner338 NUMERICAL METHODs 6 Benne o 1 dx , 5 5 Evaluate Lyx? , using Trapezoidal rule with h = 0.2 Hence determine the va| TApril 92] 7 1 Here h = 0.2. The values of the function y = 7753 for each point of subdivision are given below. x 0 0.2 0.4 0.6 08 1 _-1 1 | 0.9615 | 0.8621 | 0.7353 | 0.6098 | 0.5 te? | yo | 2 Ys ya | Ys By Trapezoidal rule we have, 1 dx h S7Ae = 7 (Ooty) +201 ty. +93 +90] o 0.2 = U5 +2 (3.1687)]_ = (0.1) (7.8374) 1 J aes = 0.78374 0 We know that 1d 1 = Gani gp! ee dx. Te = (an! x)) =G sat! ee =_4 (0.78374) [From Trapezoidal Rule] mn = 3.13496 o ample 4 iy aes 2 i 7 2 Using Trapezoidal rule evaluate | y dx from the following table, om x | 06 { 08 | 10 [12 [14 | 16 | 18 | 20 y | 1.23 | 1.58 | 2.03 | 4.32 | 6.25 | 8.36 | 10.23 | 1245 Here h = 0.2 mS UNITS —_ Scanned with CamScannerer NUMERICAL DIFFERENTIATION AND INTEGRATION 3.37 r=] 06 | 08 | 10 [12 [14] 16] 18720 1.23 | 1.58 | 2.03 | 4.32 | 6.25 | 8.36 | 10.23 | 1245 yo | 1 | % | ys | ym | 5 | 6 | 9 By Trapezoidal rule, we have, 2 h Jyde= 7 (ot +201 +¥9 +93 +4 +95 +y—)] 0.6 0.2 = “7 113.68 +2 (1.584 2.03 + 4,3246.2548.36+10.23)] = 0.1) [79.22] 2 Jyde = 7.922 06 Q SIMPSON’S 3 RULE Putting n = 2 in Pa above relation (A) (Refer Pg, No. 3.32) and pins all differences above the second we get, 22 yaya = i [oy +2 Fan r3(S - >) aro Po 670+ 6Ayp + Ayo ail + Ayo + Lay ]- [So Seee en " Gyo + 6 (1 ~ Yo) + ¥2 — 271 + YO Bone TE h 3 cee 30+ 2h 2p yG@)dr = u Ly0+4y1 +92] erie *0 Similarly for the next two intervals Xq + 2h to xo + 4h we get, 20+ 4h" h 2) J y@de = 3 [y+ 4y3 +94] a 3y+2h . UNITS Scanned with CamScannerjjuMERIGAL DIFFERENTIATION AND INTEGRATION 3.44 ee aeSATON 8A ht < -799 (b-a) 5 1 :, Error in the Simpson’s 3 rule is of the order h4, (OM Example 1 iy S2 Compute the value of the definite integral f log.x dx or 4 52 we [Inxdx using Simbsd s rule, ° ye fe ie ‘ Se weeps We scion fs Divide the interval of integration into six equal ie oct of width 0.2 i.e., h = 0.2. The values of the function y = In x are next calculated for each point of subdivision as given below. x | 40 4.2 44 46 48 5.0 5.2. In x| 1.386294 | 1.435084 | 1.481604 | 1.526056 | 1.568616 | 1.609437 | 1.648658 10 y 22 33 v4 JS. 6, By Simpson’s rule, we have 32 h J Inxdx =3 [00 + ye) +2 02 + y4) +401 +93 + y5)] 4 = a2 [3.034952 + 2 (3.050221)-+ 4 (4.570577)] - 5. Inx dx 7 : | Evaluate. f ro? de by dividing the range of integration into 4 0 sin ec” using Simpson’s rule. [Nov. ’91, Nov. '89] 1-0 Here the length of the interval is h =—g~ = 0.25. The a ofthe function y = e~ for each point of subdivision are given OW. oe UNITS Scanned with CamScannerNUMERICAI aaa 0 [= [0 | 025 0.5 0.75 1 ex |t 0.9394 | 0.7788 | 0.5698 | 0.3678 Yo YI y2. 3 M4 By Simpson’s rule we have feta = (oot +2nt4Qr ty] 0 _ 0 1 mM a 2, fede = 0.7468 f 0 6 BennrE é ” impson’s Find the value of { \1—0.162 sin’ x dx, using Simpson's 0 one third rule. Let us divide the interval of integration into 6 equil subintervals i -M2-0_ mn _ her h = 65, 212, The values of the function y =-/1—0.162sin'x for each point of subdivisions are given below. {0 a lea [ose ee oe |e 12 6 4 3 12 2 y | 1.0000 | 0:9946 | 0.9795 | 0.9586 | 0.9373 | 0.9215 | 0.915 yo. Yn v2 Y3 v4 5 % . 1 By Simpson’s 3 Tule, we have / 7 h + | Jy de =3 [00+ 96) +4 (1 ty3-+95) + 2024" 0 mUNITS... -_ Scanned with CamScanner= NUMERICAL DIFFERENTIATION AND INTEGRATION 3.43 (hahaa *P T= O62 sin x de = 35 | (1.0000 + 0.9154)+ 400.9946 0 +.0.9586 + 0.9213) + 2(0.9795 + 0.9373) } Pirodersin’ x dx= 1.5051 0 4 b L 3 1 1 Find the value of log 2 from es dx using Simpson’s 3 rule with h = 0.25. [April ’91] 2 Given h = 0.25. The values of the function y= 74 for each point of subdivisions are given below. ¥ 0 0.25 0.5 0.75 1.0 7 0 | 0.06154 | 0.22222 | 0.39560 | 0.5000 >=1e8 | yo | 22 %3 ¥4 tule, we have i 3 h 3 (Oot ya) + 224401 +y3)] = 925 (0105) +2.(0.22222) +4 (0.06154 + 0.39560) 925 tos + 0.40444 + 1.82856) , its dx= 0.231083 We know that, — a Scanned with CamScannerNUMERICAL METHOns Be ae = § (log +2909 1 = 4 (og 2- log 1) = 3 log,2 1 1 x2 x vlog? = Ses dx 1 log 2? _= 0.231083 6 Bere + When a train is moving at 30 metres per second steam is shut off and brakes are applied. The speed of the train (V) in metres per second after t seconds is given by t| 0 5 | 10 | 15 | 20| 25 | 30 | 35 | 40 V | 30 | 24 |19.5| 16 | 13.6|11.7| 10.0| 8&5 | 7.0 Using Simpson’s rule determine the distance moved by the train in 40 secs. We know that velocity is the rate of change displacement. . ds ie, V = Gy ods=Vdt Here we have to find the total distance moved by the train in 40 secs. 40 = fVd 0 The given table is [+ {o | 5 | 10] 15 v | 30 | 24 | 19.5] 16 | Yo | Yi | Va | V3 13.6 V4 17 ‘By Simpson's rule we have 40 S= [Vd afl Scanned with CamScanneree \ \ IFFERENTIATION AND INTEGRATION NUMERICAL DI 3.45 a” B [Vg + Ve) +22 + V4 4 Vo) +4 (Vy + V5 4 V5 + VDI i 3p7 +2095 +13.6+10.0)+ 4024416 +11.748.5)] 2 3 [37 + 86.2 +240.8] = 606.66 metres, = Distance moved by the train in 40 secs = 606.66 m. ¢ Barney 3s Given €9 = 1, e! = 2.72, e? = 7.39, e3 = 20.09, e4 = 54.60. Use 4 Simpson’s rule to find an approximate value of { & dx. Also compare your result with the exact value of the integral. (AMIE, S'°88] The given values can be arranged in the form of table as given below. ee z 0 1 2 3 4 ye 1 2.72 7.39 | 20.09 | 54.60 Yo y! Y2 ¥3 Ya. By Simpson’s rule, # A J ede 3 [00 +4) +292 +4 01 +y3)] we have a sceaetenetie 1 = 5 155.604 14.78 +4 (2.72 +2009)) 4 [70.38 + 91.24] Now by ordinary integration we get 4 Jetde = (e9) = ete = 54.598-1 0 4 Jetde = 53.598 decseetesseeecereeee UNITS Scanned with CamScannerstan NUMERICAL METHODs 6 EE + A river is 80 feet wide. The depth ‘d’ in feet at a distance x feet from one bank is given by the following table: x [0 | 10 | 20 | 30 | 40 | 50 | 60 | 70 | a Seer be a bas bet ep Find approximately the area of cross section of the river using Simpson’s rule. [AMIE S’ 76] Here h = 10. The given table is x | 0 | 10 | 20 | 30 | 40 | 50 | 60 | 70 | 80 i oj; 4 ]7)]9 | 12] 15] 14] 8] 3 } Yo_| Yi | Y2 | Ys | 4 | Ys | Yo | Y7 | YM yed By Simpson’s rule, we have 80 Area of cross-section = | y dx 8 i A 3 [00+ yg) +2 (2 +4 +96) +4 1 +3 +5 + ID | 1 2 [3 + 2 (33) +4 (36)} Area of cross section = 710 sq. feet. CH Example 38 RY eee \ 14 Evaluate | (sinx~Inx + e%) dx by ‘Simpson rs4 rule, \ 02 Let us divide the interval of integration i parts ‘ gration into twelve equal pa by faking h = 0.1. Now the table of values of the given function 2 sin In:t+ eat each point of subdivision is as given blo™— *|_02 | 03 | 04 [ 05 | 06 | 07 | % ¥) 3.02951 |2.84936 2.79754 | 2.82130 |2.80754 | 3.01465 | 3.166 * {| m |» | ys | wm | 9s x| 0.9 1.0 Ll 1.2 13 14 ¥/ 3.34830) 3.55935 3.80007) 4.06084 |4.37050| 4.70418 | | 1% %8 9 Yio | yn y2 BUNIT3.,. Scanned with CamScanner— N Y yuMERIGAL DIFFERENTIATION AND INTEGRATION EES ’”““ OE 4 rg By Simpson’s 3 rule, we have " A = 3 [00+ Y12) +2 02 +4 +9643 +40) + 401 Ol “3 (7.73369 + 2 (16.4907) + 4 (20.20418)] 4.05106 3.47 +93 +95 +7 + 99+ NI , 02 14 » [Ginx-Inx +e*) dx = 4.05106 4 Bere s 5 \ Use Simpson sf rule to estimate the value of | f (x) dx given ' 1 x 1 2 3 4 5 y=f@)| 13 50 70 so | 100 ei Yo ni 2 ys [ va_| ' | a 4 By Simpson’s : rule, we have 5 J f@) dx =F roo+y +200 +4 014991 1 5 { f@)de = 257.67 1 [.(13 + 100) + 2 (70) + 4 (50 + 80)] [113 + 140 +520] seo UNITS Scanned with CamScannerag NUMERICAL METHODS. 6 Bence & 4 Evaluate § f (x) dx from the following table by Simpson's > 1 rule, x 1 2 3 4 u 1 8 27 6 y=FO) | yo yn Jn Ys By Simpson’s 3 tule, we have ‘ 3h J FQ) ax =“ LO +y3) +3 01 +92) 1 1 311 43(8)4+3 0 +64) A [1424481 +64) = 1170) 4 J f@) de = 63.75 1 (Example 11 KY m2 3 Evaluate | sin x dx, using Simpson’s g tule. 0 To use Simpson’s 3 rule the number of subintervals shou! be a multiple of 3. Hence we divide the interval of integration ny). (0, 5) into 9 subintervals of width 8 . Let y= sin x. The values of the function y = sin x for each point of subdivisions are sv" below. = *)o,a [au [us x | & [x wi] w |e |e |e |e |e |e ¥ | © | 0.1736 | 0.3420 | 0.5000 | 0.628 | 0.7600 | 0.8660 | 0.9397 al Yo} “1 | 2 | | wm | xs | x | 9 | % UNITS... Scanned with CamScannereS uw RIGAL DIFFERENTIATION AND INTEGRATION 3.49 impson’s® rule, we hav By Simpson’s g rule, e 3h *g [ot yo) +3 Cy +92 +Y4 + YstY7 +8) = & u +203 +96) ] w2 ». Psinxde = FO + 1) +3 (0.1736 + 0.3420 + 0.6428 0 + 0.7660 + 0.9397 + 0.9848) + 2(0.5 + 0.8660) ] 3m (15.2787) Jsinxde = 0.999988 0 Checking: { sinxdx = [ -cosx]"?=1. 0 & é The velocity V of a particle at distances from a point on its path is given by the table : S(feet) | 0 | 10 | 20| 30 | 40| 50 | 60 V (feet/sec) | 47 | 58 | 64 | 65 | 61 | 52 | 38) Estimate the time taken to travel 60 feet by using Simpson’s one-third rule. Compare the result with Simpson’s Z rule. ‘ [AMIE S’ 90] We know that the rate of change of displacement is velocity. , a _ he, Gy = Vv (or) ds = Vdt ie, dt = ads + (1) i Here we want to find the time taken to travel 60 feet. Therefore r 60 Oy integrate (1) from 0 to 60, we get | dt= iy ds 0 0 UNITS Scanned with CamScannerso NUMERICAL METHOD The time taken to travel 60 feet is 60) 60 t= fyds= Jyd 0 0 The given table can be written as given below. x(=s)| 0 10 20 30 40 50 60 10.02127|0.01723|0.01563] 0.01538} 0.01639 |0.01923|0.0263| yo | M1 yo | 93 4 ys | % By Simpson’s one third rule, we have 60 J 0 ney h yds = 3 [0% +¥6) +202 +94) +401 +93 495) = 2 [(0.02127 + 0.0263) + 2 (0.01563 + 0.01639) +4 (0.01724 + 0.01538 + 0.01923)] a 2 [0.04757 + 0.06404 + 0.2074] = 1.063 secs. Hence time taken to travel 60 feet is 1.063 secs. By Simposon’s 3 rule 3h =F loo tye +301 +92 +94 499) +209) " 3x10 “g__ ((0.02127 + 0.02630) +3 (0.01723 + 0.01563 + 0.01639 + 0.0! +2.001538) 3.75 (0.04757 + 0.20544 + 0.03076] 923) S = " 1.064 secs. mUNITS A Scanned with CamScannerPer, nuMERICAL DIFFERENTIATION AND INTEGRATION > [Example 13] $ By di widing the range into ten equal parts, evaluate Tsin xdx 0 i , iy using Simpson’s 3 rule. Is it possible to evaluate the same by th Simpson 3 rule, Justify your answer. Here range =7—-0 = 0 & 4= 10 The values of the function y = sin x for each point of subdivisions are given below. yo, a | & |e | & [eR] & [a |e |e |e 6 10 10 to | 10 10 To 0 10 FL 0 | oso | ose7s [ans [oss | 1.0) oss | ooo] osere | 03050 | 0 % Ea D2 ¥3, 4 ¥s % ” 8 %9 v0 1 By Simpson’s 3 rule zr h { sinxdx =3[00+y10) +2 (2+ ¥4+Y6+ Ys) 0 +401 +93 +95 +97 +991 : = 10 =3 (+0) +2 (0.5878 + 0.9511 +0.9511 + 0.5878) + 4 (0.3090 + 0.8090 + 1,0 + 0.8090 + 0.3090] x J sinx dy = 2.00091 0 : 3th = Note: Here we cannot use Simpson's g rule since the subintervals is not a multiple of 3. UNITS Scanned with CamScanner