1

CHAPTER 20
20.1 A plot of log10k versus log10f can be developed as
0.7 0.2 0 -0.2 -0.4 -0.6 y = 0.4224x - 0.83 R2 = 0.9532 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3

As shown, the best fit line is log10 f = 0.422363 log10 k 0.83 Therefore, 2 = 10 0.83 = 0.147913 and 2 = 0.422363, and the power model is y = 0.147913x 0.422363 The model and the data can be plotted as
1.5 1 0.5 0 0 20 40 60 80 100 120 140 160 y = 0.1479x0.4224 R2 = 0.9532

20.2 We can first try a linear fit

1800 1600 1400 1200 -60 -30 0 30 60 90 120 y = 2.5826x + 1365.9 R2 = 0.9608

As shown, the fit line is somewhat lacking. Therefore, we can use polynomial regression to fit a parabola

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

1800 1600 1400 1200 -60 -30 0

y = 0.0128x + 1.8164x + 1331 R = 0.9934

2

30

60

90

120

This fit seems adequate in that it captures the general trend of the data. Note that a slightly better fit can be attained with a cubic polynomial, but the improvement is marginal. 20.3 This problem is ideally suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible. x0 = 20 x1 = 15 x2 = 25 x3 = 10 x4 = 30 x5 = 5 x6 = 0 f(x0) = 8.17 f(x1) = 9.03 f(x2) = 7.46 f(x3) = 10.1 f(x4) = 6.85 f(x5) = 11.3 f(x6) = 12.9

The results of applying Newtons polynomial at T = 18 are

Order 0 1 2 3 4 5 6 f(x) 8.17000 8.51400 8.49600 8.49264 8.49286 8.49125 8.48827 Error 0.344 -0.018 -0.00336 0.00022 -0.00161 -0.00298

The minimum error occurs for the third-order version so we conclude that the interpolation is 8.4926. 20.4 We can use MATLAB to solve this problem,
>> >> >> >> format long T = [0 5 10 15 20 25 30]; c = [12.9 11.3 10.1 9.03 8.17 7.46 6.85]; p = polyfit(T,c,3) 0.00652380952381 -0.34111111111111

p = -0.00006222222222 12.88785714285715

Thus, the best-fit cubic would be

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

3 c = 12.8878571 0.34111111T + 0.00652381T 2 0.000062222T 3 We can generate a plot of the data along with the best-fit line as
>> Tp=[0:30]; >> cp=polyval(p,Tp); >> plot(Tp,cp,T,c,'o')

We can use the best-fit equation to generate a prediction at T = 8 as

>> y=polyval(p,8) y = 10.54463428571429

20.5 The multiple linear regression model to evaluate is o = a 0 + a1T + a 2 c As described in Section 17.1, we can use general linear least squares to generate the best-fit equation. The [Z] and y matrices can be set up using MATLAB commands as
>> format long >> t = [0 5 10 15 20 25 30]; >> T = [t t t]'; >> c = [zeros(size(t)) 10*ones(size(t)) 20*ones(size(t))]'; >> Z = [ones(size(T)) T c]; >> y = [14.6 12.8 11.3 10.1 9.09 8.26 7.56 12.9 11.3 10.1 9.03 8.17 7.46 6.85 11.4 10.3 8.96 8.08 7.35 6.73 6.2]';

The coefficients can be evaluated as

>> a=inv(Z'*Z)*(Z'*y)
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

4
a = 13.52214285714285 -0.20123809523809 -0.10492857142857

Thus, the best-fit multiple regression model is o = 13.52214285714285 0.20123809523809T 0.10492857142857c We can evaluate the prediction at T = 17 and c = 5 and evaluate the percent relative error as
>> op = a(1)+a(2)*17+a(3)*5 op = 9.57645238095238

We can also assess the fit by plotting the model predictions versus the data. A one-to-one line is included to show how the predictions diverge from a perfect fit.
16

12

4 4 8 12 16

The cause for the discrepancy is because the dependence of oxygen concentration on the unknowns is significantly nonlinear. It should be noted that this is particularly the case for the dependency on temperature. 20.6 The multiple linear regression model to evaluate is o = a 0 + a1T + a 2T 2 + a 3T 3 + a 4 c The Excel Data Analysis toolpack provides a convenient tool to fit this model. We can set up a worksheet and implement the Data Analysis Regression tool as

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

When the tool is run, the following worksheet is generated

Thus, the best-fit model is o s = 14.027143 0.336423T + 0.00574444T 2 0.000043704T 3 0.10492857c The model can then be used to predict values of oxygen at the same values as the data. These predictions can be plotted against the data to depict the goodness of fit.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

Thus, although there are some discrepancies, the fit is generally adequate. Finally, the prediction can be made at T = 20 and c = 10, o s = 14.027143 0.336423(20) + 0.00574444(20) 2 0.000043704( 20) 3 0.10492857(10) = 8.19754 which compares favorably with the true value of 8.17 mg/L. 20.7 (a) The linear fit is

80 60 40 20 0 0

y = 1.05897x + 0.81793 R2 = 0.90583

20

40

60

80

The tensile strength at t = 32 can be computed as y = 1.05897(32) + 0.81793 = 34.7048913 (b) A straight line with zero intercept can be fit as

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

80 60 40 20 0 0

y = 1.07514x R2 = 0.90556

20

40

60

80

For this case, the tensile strength at t = 32 can be computed as y = 1.07514(32) = 34.40452 20.8 Linear regression with a zero intercept gives [note that T(K) = T(oC) + 273.15].
15000 10000 5000 0 0 100 200 300 400 500 y = 29.728x R2 = 0.9999

Thus, the fit is p = 29.728T Using the ideal gas law p V R= T n For our fit p = 29.728 T For nitrogen, n= 1 kg 28 g/mole

Therefore,
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

10 R = 29.728 3 = 8.324 10 / 28 This is close to the standard value of 8.314 J/gmole. 20.9 This problem is ideally suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible. x0 = 740 x1 = 760 x2 = 720 x3 = 780 x4 = 700 f(x0) = 0.1406 f(x1) = 0.15509 f(x2) = 0.12184 f(x3) = 0.16643 f(x4) = 0.0977

The results of applying Newtons polynomial at T = 750 are

Order 0 1 2 3 4 f(x) 0.14060 0.14785 0.14838 0.14831 0.14831 Error 0.007245 0.000534 -7E-05 0.00000

Note that the third-order polynomial yields an exact result, and so we conclude that the interpolation is 0.14831. 20.10 A program can be written to fit a natural cubic spline to this data and also generate the first and second derivatives at each knot.
Option Explicit Sub Splines() Dim i As Integer, n As Integer Dim x(100) As Double, y(100) As Double, xu As Double, yu As Double Dim dy As Double, d2y As Double Dim resp As Variant Range("a5").Select n = ActiveCell.Row Selection.End(xlDown).Select n = ActiveCell.Row - n Range("a5").Select For i = 0 To n x(i) = ActiveCell.Value ActiveCell.Offset(0, 1).Select y(i) = ActiveCell.Value ActiveCell.Offset(1, -1).Select Next i Range("c5").Select Range("c5:d1005").ClearContents For i = 0 To n Call Spline(x(), y(), n, x(i), yu, dy, d2y) ActiveCell.Value = dy PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

9
ActiveCell.Offset(0, 1).Select ActiveCell.Value = d2y ActiveCell.Offset(1, -1).Select Next i Do resp = MsgBox("Do you want to interpolate?", vbYesNo) If resp = vbNo Then Exit Do xu = InputBox("z = ") Call Spline(x(), y(), n, xu, yu, dy, d2y) MsgBox "For z = " & xu & Chr(13) & "T = " & yu & Chr(13) & _ "dT/dz = " & dy & Chr(13) & "d2T/dz2 = " & d2y Loop End Sub Sub Spline(x, y, n, xu, yu, dy, d2y) Dim e(100) As Double, f(100) As Double, g(100) As Double, r(100) As Double, d2x(100) As Double Call Tridiag(x, y, n, e, f, g, r) Call Decomp(e(), f(), g(), n - 1) Call Substit(e(), f(), g(), r(), n - 1, d2x()) Call Interpol(x, y, n, d2x(), xu, yu, dy, d2y) End Sub Sub Tridiag(x, y, n, e, f, g, r) Dim i As Integer f(1) = 2 * (x(2) - x(0)) g(1) = x(2) - x(1) r(1) = 6 / (x(2) - x(1)) * (y(2) - y(1)) r(1) = r(1) + 6 / (x(1) - x(0)) * (y(0) - y(1)) For i = 2 To n - 2 e(i) = x(i) - x(i - 1) f(i) = 2 * (x(i + 1) - x(i - 1)) g(i) = x(i + 1) - x(i) r(i) = 6 / (x(i + 1) - x(i)) * (y(i + 1) - y(i)) r(i) = r(i) + 6 / (x(i) - x(i - 1)) * (y(i - 1) - y(i)) Next i e(n - 1) = x(n - 1) - x(n - 2) f(n - 1) = 2 * (x(n) - x(n - 2)) r(n - 1) = 6 / (x(n) - x(n - 1)) * (y(n) - y(n - 1)) r(n - 1) = r(n - 1) + 6 / (x(n - 1) - x(n - 2)) * (y(n - 2) - y(n - 1)) End Sub Sub Interpol(x, y, n, d2x, xu, yu, dy, d2y) Dim i As Integer, flag As Integer Dim c1 As Double, c2 As Double, c3 As Double, c4 As Double Dim t1 As Double, t2 As Double, t3 As Double, t4 As Double flag = 0 i = 1 Do If xu >= x(i - 1) And xu <= x(i) Then c1 = d2x(i - 1) / 6 / (x(i) - x(i - 1)) c2 = d2x(i) / 6 / (x(i) - x(i - 1)) c3 = y(i - 1) / (x(i) - x(i - 1)) - d2x(i - 1) * (x(i) - x(i - 1)) / 6 c4 = y(i) / (x(i) - x(i - 1)) - d2x(i) * (x(i) - x(i - 1)) / 6 t1 = c1 * (x(i) - xu) ^ 3 t2 = c2 * (xu - x(i - 1)) ^ 3 t3 = c3 * (x(i) - xu) t4 = c4 * (xu - x(i - 1)) yu = t1 + t2 + t3 + t4 t1 = -3 * c1 * (x(i) - xu) ^ 2 t2 = 3 * c2 * (xu - x(i - 1)) ^ 2 PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

10
t3 = -c3 t4 = c4 dy = t1 + t2 + t3 + t4 t1 = 6 * c1 * (x(i) - xu) t2 = 6 * c2 * (xu - x(i - 1)) d2y = t1 + t2 flag = 1 Else i = i + 1 End If If i = n + 1 Or flag = 1 Then Exit Do Loop If flag = 0 Then MsgBox "outside range" End End If End Sub Sub Decomp(e, f, g, n) Dim k As Integer For k = 2 To n e(k) = e(k) / f(k - 1) f(k) = f(k) - e(k) * g(k - 1) Next k End Sub Sub Substit(e, f, g, r, n, x) Dim k As Integer For k = 2 To n r(k) = r(k) - e(k) * r(k - 1) Next k x(n) = r(n) / f(n) For k = n - 1 To 1 Step -1 x(k) = (r(k) - g(k) * x(k + 1)) / f(k) Next k End Sub

Here is the output when it is applied to the data for this problem:

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

11

The plot suggests a zero second derivative at a little above z = 1.2 m. The program is set up to allow you to evaluate the spline prediction and the associated derivatives for as many cases as you desire. This can be done by trial-and-error to determine that the zero second derivative occurs at about z = 1.2315 m. At this point, the first derivative is 73.315 oC/m. This can be substituted into Fouriers law to compute the flux as J = 0.02 cal C m cal = 0.01466 73.315 s cm C m 100 cm cm 2 s

20.11 This is an example of the saturated-growth-rate model. We can plot 1/[F] versus 1/[B] and fit a straight line as shown below.
0.1 0.08 0.06 0.04 0.02 0 0 2 4 6 8 10 y = 0.008442x + 0.010246 R2 = 0.999932

The model coefficients can then be computed as

3 =

1 = 97.60128 0.010246

3 = 97.60128(0.008442) = 0.823968
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

12 Therefore, the best-fit model is [ B] = 97.60128 [F ] 0.823968 + [ F ]

A plot of the original data along with the best-fit curve can be developed as
120 100 80 60 40 20 0 0 2 4 6 8 10

20.12 Nonlinear regression can be used to estimate the model parameters. This can be done using the Excel Solver

Here are the formulas:

Therefore, we estimate k01 = 7,526.842 and E1 = 4.478896.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

13 20.13 Nonlinear regression can be used to estimate the model parameters. This can be done using the Excel Solver. To do this, we set up columns holding each side of the equation: PV/(RT) and 1 + A1/V + A2/V. We then form the square of the difference between the two sides and minimize this difference by adjusting the parameters.

Here are the formulas that underly the worksheet cells:

Therefore, we estimate A1 = -237.531 and A2 = 1.192283. 20.14 The standard errors can be computed via Eq. 17.9 sy/ x = n = 15
Sr Number of model parameters fit (p) sy/x Model A 135 2 3.222517 Model B 105 3 2.95804 Model C 100 5 3.162278

Sr n p

Thus, Model B seems best because its standard error is lower. 20.15 A plot of the natural log of cells versus time indicates two straight lines with a sharp break at 2. The Excel Trendline tool can be used to fit each range separately with the exponential model as shown in the second plot.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

14

2 1 0 -1 0 -2 -3 2 4 6

6 4 2 0 0 2 4 6 8 y = 0.1000e
2 1.1999x

R = 1.0000

y = 0.4951e
2

0.4001x

R = 1.0000

20.16 This problem can be solved with Excel,

20.17 This problem can be solved with a power model, A = a 0 H a1 W a2 This equation can be linearized by taking the logarithm log10 A = log10 a 0 + a1 log10 H + a 2 log10 W This model can be fit with the Excel Data Analysis Regression tool,

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

15

The results are

Therefore the best-fit coefficients are a0 = 10 1.86458 = 0.013659, a1 = 0.5218, and a2 = 0.5190. Therefore, the model is A = 0.013659 H 0.5218W 0.5190 We can assess the fit by plotting the model predictions versus the data. We have included the perfect 1:1 line for comparison.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

16

2.3

2.1

1.9 1.9 2.1 2.3

The prediction at H = 187 cm and W = 78 kg can be computed as A = 0.013659(187) 0.5218 (78) 0.5190 = 2.008646 20.18 The Excel Trend Line tool can be used to fit a power law to the data:

Note that although the model seems to represent a good fit, the performance of the lowermass animals is not evident because of the scale. A log-log plot provides a better perspective to assess the fit:
1000 100 10 1 0.1 0.1 1 10 100 1000 y = 3.3893x
2 0.7266

R = 0.9982

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

17

20.19 For the Casson Region, linear regression yields

= 0.065818 + 0.180922

r 2 = 0.99985

For the Newton Region, linear regression with zero intercept gives

= 0.032447

r 2 = 0.99992

On a Casson plot, this function becomes

= 0.180131
We can plot both functions along with the data on a Casson plot

4
3 2 1 0 0

= 0 .065818 + 0.180922

= 0 .180131
5 10 15 20

20.20 Recall from Sec. 4.1.3, that finite divided differences can be used to estimate the derivatives. For the first point, we can use a forward difference (Eq. 4.17) d 96.6 87.8 = = 0.1725 d 204 153 For the intermediate points, we can use centered differences. For example, for the second point d 176 87.8 = = 0.8647 d 255 153 For the last point, we can use a backward difference d 4258 3380 = = 17.2157 d 765 714 All the values can be tabulated as
87.8 96.6 176 263 153 204 255 306
d /d 0.1725 0.8647 1.6314 1.7157

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

18
351 571 834 1229 1624 2107 2678 3380 4258 357 408 459 510 561 612 663 714 765 3.0196 4.7353 6.4510 7.7451 8.6078 10.3333 12.4804 15.4902 17.2157

We can plot these results and after discarding the first few points, we can fit a straight line as shown (note that the discarded points are displayed as open circles).
20 15 10 5 0 0 1000 2000 3000 4000 y = 0.003557x + 2.833492 R2 = 0.984354

Therefore, the parameter estimates are Eo = 2.833492 and a = 0.003557. The data along with the first equation can be plotted as
12000 8000 4000 0 0 200 400 600 800

As described in the problem statement, the fit is not very good. We therefore pick a point near the midpoint of the data interval ( = 612, = 2107) We can then plot the second model 2107 0.003557 = 0.003557 ( 612 ) 1) = 269.5(e 0.003557 1) ( e 1 e The result is a much better fit
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

19

5000 4000 3000 2000 1000 0 0 100 200 300 400 500 600 700 800

20.21 The problem is set up as the following Excel Solver application. Notice that we have assumed that the model consists of a constant plus two bell-shaped curves: f ( x) = c + k1e k1 ( x a1 )
2 2

k 2 e k 2 ( x a2 )

The resulting solution is

Thus, the retina thickness is estimated as 0.312 0.24 = 0.072. 20.22 Simple linear regression can be applied to yield the following fit

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

20

100 80 60 40 20 0 0

y = 7.6491x - 1.5459 R2 = 0.8817

10

12

The shear stress at the depth of 4.5 m can be computed as

= 7.6491(4.5) 1.5459 = 32.875

20.23 (a) and (b) Simple linear regression can be applied to yield the following fit
4 3 2 1 0 0 0.5 1 1.5 2 2.5 3 3.5

y = 0.7335x + 0.7167 R2 = 0.8374

(c) The minimum lane width corresponding to a bike-car distance of 2 m can be computed as y = 0.7335( 2) + 0.7167 = 2.1837 m 20.24 (a) and (b) Simple linear regression can be applied to yield the following fit
24 20 16 12 80 90 100 110 120 130 140 150 y = 0.1519x + 0.8428 R2 = 0.894

(c) The flow corresponding to the precipitation of 120 cm can be computed as Q = 0.1519(120) + 0.8428 = 19.067 (d) We can redo the regression, but with a zero intercept
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

21

24 20 16 12 80

y = 0.1594x R2 = 0.8917

90

100

110

120

130

140

150

Thus, the model is Q = 0.1594 P where Q = flow and P = precipitation. Now, if there are no water losses, the maximum flow, Qm, that could occur for a level of precipitation should be equal to the product of the annual precipitation and the drainage area. This is expressed by the following equation. cm Qm = A( km 2 ) P yr For an area of 1100 km2 and applying conversions so that the flow has units of m3/s cm 10 6 m 2 1 m d yr Qm = 1,100 km 2 P yr km 2 100 cm 86,400 s 365 d Collecting terms gives Qm = 0.348808P Using the slope from the linear regression with zero intercept, we can compute the fraction of the total flow that is lost to evaporation and other consumptive uses can be computed as F= 0.348808 0.1594 = 0.543 0.348808

20.25 The data can be computed in a number of ways. First, we can use linear regression.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

22

12 10 8 6 4 2 0 0

y = 0.2879x - 0.629 R = 0.9855

2

10

15

20

25

30

35

40

For this model, the chlorophyll level in western Lake Erie corresponding to a phosphorus concentration of 10 mg/m3 is c = 0.2879(10) 0.629 = 2.2495 One problem with this result is that the model yields a physically unrealistic negative intercept. In fact, because chlorophyll cannot exist without phosphorus, a fit with a zero intercept would be preferable. Such a fit can be developed as
12 10 8 6 4 2 0 0 5 10 15 20 25 30 35 40 y = 0.2617x R = 0.9736
2

For this model, the chlorophyll level in western Lake Erie corresponding to a phosphorus concentration of 10 mg/m3 is c = 0.2617(10) = 2.617 Thus, as expected, the result differs from that obtained with a nonzero intercept. Finally, it should be noted that in practice, such data is usually fit with a power model. This model is often adopted because it has a zero intercept while not constraining the model to be linear. When this is done, the result is

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

23

12 10 8 6 4 2 0 0

y = 0.1606x1.1426 R2 = 0.9878

10

15

20

25

30

35

40

For this model, the chlorophyll level in western Lake Erie corresponding to a phosphorus concentration of 10 mg/m3 is c = 0.1606(10)1.1426 = 2.2302 20.26 m= 4.6 = 0.46 10 n= 14 = 1.4 10

Use the following values x0 = 0.3 x1 = 0.4 x2 = 0.5 x3 = 0.6 f(x0) = 0.08561 f(x1) = 0.10941 f(x2) = 0.13003 f(x3) = 0.14749

MATLAB can be used to perform the interpolation >> format long >> x=[0.3 0.4 0.5 0.6]; >> y=[0.08561 0.10941 0.13003 0.14749]; >> a=polyfit(x,y,3) a= Columns 1 through 3 0.00333333333331 -0.16299999999997 0.35086666666665 Column 4 -0.00507000000000 >> polyval(a,0.46) ans = 0.12216232000000
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

24

This result can be used to compute the vertical stress 100 = 1.552795 4.6(14) z = 1.552795(0.1221623) = 0.189693 q= 20.27 This is an ideal problem for general linear least squares. The problem can be solved with MATLAB:
>> >> >> >> a = 4.0046 2.9213 1.5647 t=[0.5 1 2 3 4 5 6 7 9]'; p=[6 4.4 3.2 2.7 2.2 1.9 1.7 1.4 1.1]'; Z=[exp(-1.5*t) exp(-0.3*t) exp(-0.05*t)]; a=inv(Z'*Z)*(Z'*p)

Therefore, A = 4.0046, B = 2.9213, and C = 1.5647. 20.28 First, we can determine the stress

25000 = 2,347.418 10.65

We can then try to fit the data to obtain a mathematical relationship between strain and stress. First, we can try linear regression:
0.01 0.008 0.006 0.004 0.002 0 -0.002 0 2000 4000 6000 8000 y = 1.37124E-06x - 2.28849E-03 R2 = 0.856845

This is not a particularly good fit as the r2 is relatively low. We therefore try a best-fit parabola,

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

25

0.01 0.008 0.006 0.004 0.002 0 0

y = 2.8177E-10x2 - 8.4598E-07x + 1.0766E-03 R2 = 0.98053

2000

4000

6000

8000

We can use this model to compute the strain as

= 2.8177 10 10 (2347.4178) 2 8.4598 10 7 (2347.4178) + 1.0766 10 3 = 6.4341 10 4

The deflection can be computed as L = 6.4341 10 4 (9) = 0.0057907 m 20.29 Clearly the linear model is not adequate. The second model can be fit with the Excel Solver:

Notice that we have reexpressed the initial rates by multiplying them by 1 106. We did this so that the sum of the squares of the residuals would not be miniscule. Sometimes this will lead the Solver to conclude that it is at the minimum, even though the fit is poor. The solution is:

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

26

Although the fit might appear to be OK, it is biased in that it underestimates the low values and overestimates the high ones. The poorness of the fit is really obvious if we display the results as a log-log plot:
1.E+02 1.E+00 1.E-02 1.E-04 1.E-06 0.01 1 100 v0 v0mod

Notice that this view illustrates that the model actually overpredicts the very lowest values. The third and fourth models provide a means to rectify this problem. Because they raise [S] to powers, they have more degrees of freedom to follow the underlying pattern of the data. For example, the third model gives:

Finally, the cubic model results in a perfect fit:

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

27

Thus, the best fit is v0 = 2.45 10 5 [ S ] 3 0.385 + [ S ] 3

20.30 As described in Section 17.1, we can use general linear least squares to generate the best-fit equation. We can use a number of different software tools to do this. For example, the [Z] and y matrices can be set up using MATLAB commands as
>> >> >> >> >> format long x = [273.15 283.15 293.15 303.15 313.15]'; Kw = [1.164e-15 2.950e-15 6.846e-15 1.467e-14 2.929e-14]'; y=-log10(Kw); Z = [(1./x) log10(x) x ones(size(x))];

The coefficients can be evaluated as

>> a=inv(Z'*Z)*(Z'*y) Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 6.457873e-020. a = 1.0e+003 * 5.18067187500000 0.01342456054688 0.00000562858582 -0.03827636718750

Note the warning that the results are ill-conditioned. According to this calculation, the best-fit model is log10 K w = 5180.67 + 13.42456 log10 Ta + 0.00562859Ta 38.276367 Ta

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

28

We can check the results by using the model to make predictions at the values of the original data
>> yp=10.^-(a(1)./x+a(2)*log10(x)+a(3)*x+a(4)) yp = 1.0e-013 * 0.01161193308242 0.02943235714551 0.06828461729494 0.14636330575049 0.29218444886852

These results agree to about 2 or 3 significant digits with the original data. 20.31 Using MATLAB:
>> >> >> >> >> >> >> >> >> t=0:2*pi/128:2*pi; f=4*cos(5*t)-7*sin(3*t)+6; y=fft(f); y(1)=[ ]; n=length(y); power=abs(y(1:n/2)).^2; nyquist=1/2; freq=n*(1:n/2)/(n/2)*nyquist; plot(freq,power);

20.32 Since we do not know the proper order of the interpolating polynomial, this problem is suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible. x0 = 1.25 f(x0) = 0.7

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

29 x1 = 0.75 x2 = 1.5 x3 = 0.25 x4 = 2 f(x1) = 0.6 f(x2) = 1.88 f(x3) = 0.45 f(x4) = 6

The results of applying Newtons polynomial at i = 1.15 are

Order 0 1 2 3 4 f(x) 0.7 0.44 0.326933 0.326112 0.337286 Error -0.26 -0.11307 -0.00082 0.011174

The minimum error occurs for the second-order version so we conclude that the interpolation is 0.3269. 20.33 Here are the results of first through fourth-order regression:
8 6 4 2 0 -2 0 0.5 1 1.5 2 2.5 y = 3.54685x - 2.57288 R = 0.78500
2

8 6 4 2 0 -2 0

y = 3.3945x2 - 4.0093x + 0.3886 R2 = 0.9988

0.5

1.5

2.5

8 6 4 2 0 -2 0

y = 0.5663x3 + 1.4536x2 - 2.1693x - 0.0113 R2 = 0.9999

0.5

1.5

2.5

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

30

8 6 4 2 0 -2 0

y = 0.88686x4 - 3.38438x3 + 7.30000x2 - 5.40448x + 0.49429 R2 = 1.00000

0.5

1.5

2.5

The 2nd through 4th-order polynomials all seem to capture the general trend of the data. Each of the polynomials can be used to make the prediction at i = 1.15 with the results tabulated below:
Order 1 2 3 4 Prediction 1.5060 0.2670 0.2776 0.3373

Thus, although the 2nd through 4th-order polynomials all seem to follow a similar trend, they yield quite different predictions. The results are so sensitive because there are few data points. 20.34 Since we do not know the proper order of the interpolating polynomial, this problem is suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible. x0 = 0.25 x1 = 0.125 x2 = 0.375 x3 = 0 x4 = 0.5 f(x0) = 7.75 f(x1) = 6.24 f(x2) = 4.85 f(x3) = 0 f(x4) = 0

The results of applying Newtons polynomial at t = 0.23 are

Order 0 1 2 3 4 f(x) 7.75 7.5084 7.804752 7.813067 7.838646 Error -0.2416 0.296352 0.008315 0.025579

The minimum error occurs for the second-order version so we conclude that the interpolation is 7.805. 20.35(a) The linear fit is

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

31

30 20 10 0 0

y = 2.8082x - 0.5922 R2 = 0.9991

10

12

The current for a voltage of 3.5 V can be computed as y = 2.8082(3.5) 0.5922 = 9.2364 Both the graph and the r2 indicate that the fit is good. (b) A straight line with zero intercept can be fit as
30 20 10 0 0 2 4 6 8 10 12 y = 2.7177x R = 0.9978
2

For this case, the current at V = 3.5 can be computed as y = 2.7177(3.5) = 9.512 20.36 The linear fit is
60 50 40 30 20 10 0 0 y = 4.7384x + 1.2685 R2 = 0.9873

10

12

Therefore, an estimate for L is 4.7384. However, because there is a non-zero intercept, a better approach would be to fit the data with a linear model with a zero intercept

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

32

60 50 40 30 20 10 0 0

y = 4.9163x R2 = 0.9854

10

12

This fit is almost as good as the first case, but has the advantage that it has the physically more realistic zero intercept. Thus, a superior estimate of L is 4.9163. 20.37 Since we do not know the proper order of the interpolating polynomial, this problem is suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible. x0 = 0.5 x1 = 0.5 x2 = 1 x3 = 1 x4 = 2 x5 = 2 f(x0) = 20.5 f(x1) = 20.5 f(x2) = 96.5 f(x3) = 96.5 f(x4) = 637 f(x5) = 637

The results of applying Newtons polynomial at i = 0.1 are

Order 0 1 2 3 4 5 f(x) 20.5 4.1 -13.66 2.324 2.324 2.324 Error -16.4 -17.76 15.984 0 0

Thus, we can see that the data was generated with a cubic polynomial. 20.38 Because there are 6 points, we can fit a 5th-order polynomial. This can be done with Eq. 18.26 or with a software package like Excel or MATLAB that is capable of evaluating the coefficients. For example, using MATLAB,
>> x=[-2 -1 -0.5 0.5 1 2]; >> y=[-637 -96.5 -20.5 20.5 96.5 637]; >> a=polyfit(x,y,5) a = 0.0000 0.0000 74.0000 -0.0000 22.5000 0.0000

Thus, we see that the polynomial is

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

33 V = 74i 3 + 22.5i 20.39 Linear regression yields

20 15 10 5 0 0 100 200 300 400 500 600 700 y = 0.0195x + 3.2895 R2 = 0.9768

The percent elongation for a temperature of 400 can be computed as % elongation = 0.0195( 400) + 3.2895 = 11.072 20.40 (a) A 4th-order interpolating polynomial can be generated as
0.6 0.55 0.5 0.45 0.4 1.5 1.7 1.9 2.1 2.3 2.5 2.7 y = 0.0182x4 - 0.1365x3 + 0.1818x 2 + 0.263x + 0.1237 R2 = 1

The polynomial can be used to compute J 1 ( 2.1) = 0.018229(2.1) 4 0.13646(2.1) 3 + 0.181771(2.1) 2 + 0.262958(2.1) + 0.1237 = 0.568304 The relative error is

t =

0.568292 0.568304 100% = 0.0021% 0.568292

Thus, the interpolating polynomial yields an excellent result. (b) A program can be developed to fit natural cubic splines through data based on Fig. 18.18. If this program is run with the data for this problem, the interpolation at 2.1 is 0.56846 which has a relative error of t = 0.0295%. A spline can also be fit with MATLAB. It should be noted that MATLAB does not use a natural spline. Rather, it uses a so-called not-a-knot spline. Thus, as shown below, although it also yields a very good prediction, the result differs from the one generated with the natural spline,
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

34

>> >> >> >>

format long x=[1.8 2 2.2 2.4 2.6]; y=[0.5815 0.5767 0.556 0.5202 0.4708]; spline(x,y,2.1)

ans = 0.56829843750000

This result has a relative error of t = 0.0011%. 20.41 The fit of the exponential model is
2500 2000 1500 1000 500 0 0 5 10 15 20 y = 97.915e0.151x R2 = 0.9996

The model can be used to predict the population 5 years in the future as p = 97.91484e 0.150992 ( 25) = 4268 20.42 The prediction using the model from Sec. 20.4 is Q = 55.9(1.23) 2.62 (0.001) 0.54 = 2.30655 The linear prediction is x0 = 1 x1 = 2 Q = 1.4 + f(x0) = 1.4 f(x1) = 8.3 8.3 1.4 (1.23 1) = 2.987 2 1

The quadratic prediction is x0 = 1 x1 = 2 x2 = 3 f(x0) = 1.4 f(x1) = 8.3 f(x2) = 24.2

24.2 8.3 8.3 1.4 32 2 1 (1.23 1)(1.23 2) = 2.19005 Q = 2.987 + 3 1

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

35

20.43 The model to be fit is S = b0 D b1 Q b2 Taking the common logarithm gives log10 S = log10 b0 + b1 log10 D + b2 log10 Q We can use a number of different approaches to fit this model. The following shows how it can be done with MATLAB,
>> >> >> >> >> >> a = -3.2563 -4.8726 1.8627 D=[1 2 3 1 2 3 1 2 3]'; S=[0.001 0.001 0.001 0.01 0.01 0.01 0.05 0.05 0.05]'; Q=[1.4 8.3 24.2 4.7 28.9 84 11.1 69 200]'; Z=[ones(size(S)) log10(D) log10(Q)]; y=log10(S); a=inv(Z'*Z)*(Z'*y)

Therefore, the result is log10 S = 3.2563 4.8726 log10 D + 1.862733 log10 Q or in untransformed format S = 0.000554 D 4.8726 Q 1.862733 (1)

We can compare this equation with the original model developed in Sec. 20.4 by solving Eq. 1 for Q, Q = 55.9897 D 2.6158 S 0.5368 This is very close to the original model. 20.44 (a) The data can be plotted as

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

36

2.0E-03 1.5E-03 1.0E-03 5.0E-04 0.0E+00 0 10 20 30 40

(b) This part of the problem is well-suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible. x0 = 5 x1 = 10 x2 = 0 x3 = 20 x4 = 30 x5 = 40 f(x0) = 1.519 103 f(x1) = 1.307 103 f(x2) = 1.787 103 f(x3) = 1.002 103 f(x4) = 0.7975 103 f(x5) = 0.6529 103

The results of applying Newtons polynomial at T = 7.5 are

Order 0 1 2 3 4 5 f(x) 1.51900E-03 1.41300E-03 1.40600E-03 1.40677E-03 1.40686E-03 1.40686E-03 Error -0.00011 -7E-06 7.66E-07 9.18E-08 5.81E-09

The minimum error occurs for the fourth-order version so we conclude that the interpolation is 1.40686 103. (b) Polynomial regression yields
2.0E-03 1.5E-03 1.0E-03 5.0E-04 0.0E+00 0 10 20 30 40 y = 5.48342E-07x2 - 4.94934E-05x + 1.76724E-03 R2 = 9.98282E-01

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

37 This leads to a prediction of

= 5.48342 10 7 (7.5) 2 4.94934 10 5 (7.5) + 1.76724 10 3 = 1.4269 10 3

20.45 For the first four points, linear regression yields
50 40 30 20 10 0 0 0.1 0.2 0.3 0.4 y = 117.16x - 1.0682 R2 = 0.9959

We can fit a power equation to the last four points.

100 80 60 40 20 0 0.38 0.39 0.4 0.41 0.42 0.43 0.44 0.45

y = 1665x3.7517 R2 = 0.9395

Both curves can be plotted along with the data.

100 80 60 40 20 0 0 0.1 0.2 0.3 0.4 0.5

20.46 A power fit to all the data can be implemented as in the following plot.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

38

100 80 60 40 20 0 0

y = 185.45x1.2876 R2 = 0.9637

0.1

0.2

0.3

0.4

0.5

Although the coefficient of determination is relatively high (r2 = 0.9637), this fit is not very acceptable. In contrast, the piecewise fit from Prob. 20.45 does a much better job of tracking on the trend of the data. 20.47 The Thinking curve can be fit with a linear model whereas the Braking curve can be fit with a quadratic model as in the following plot.
100 80 60 40 20 0 0 20 40 60 80 100 120 140 y = 5.8783E-03x 2 + 9.2063E-04x - 9.5000E-02 R2 = 9.9993E-01 y = 0.1865x + 0.0800 R2 = 0.9986

A prediction of the total stopping distance for a car traveling at 110 km/hr can be computed as d = 0.1865(110) + 0.0800 + 5.8783 10 3 (110) 2 + 9.2063 10 4 (110) 9.5000 10 2 = 91.726 m 20.48 Using linear regression gives
50 40 30 20 10 0 0 10 20 30 40 y = -0.6x + 39.75 R2 = 0.5856

A prediction of the fracture time for an applied stress of 20 can be computed as t = 0.6( 20) + 39.75 = 27.75
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

39 Some students might consider the point at (20, 40) as an outlier and discard it. If this is done, 4th-order polynomial regression gives the following fit
50 40 30 20 10 0 0 10 20 30 40 y = -1.3951E-04x4 + 1.0603E-02x3 - 2.1959E-01x2 + 3.6477E-02x + 4.3707E+01 2 R = 9.8817E-01

A prediction of the fracture time for an applied stress of 20 can be computed as t = 0.00013951(20) 4 + 0.010603(20) 3 0.21959(20) 2 + 0.036477(20) + 43.707 = 19.103 20.49 Using linear regression gives
9.9 9.8 9.7 9.6 9.5 0 20000 40000 60000 80000 100000 120000 y = -2.015000E-06x + 9.809420E+00 R2 = 9.999671E-01

A prediction of g at 55,000 m can be made as g (55,000) = 2.015 10 6 (55,000) + 9.80942 = 9.6986 Note that we can also use linear interpolation to give g 1 (55,000) = 9.698033 Quadratic interpolation yields g 2 (55,000) = 9.697985 Cubic interpolation gives g 3 (55,000) = 9.69799 Based on all these estimates, we can be confident that the result to 3 significant digits is 9.698.
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

40

20.50 A plot of log10 versus log10 can be developed as

1.2 1 0.8 0.6 y = 0.3792x + 2.0518 R2 = 0.9997 0.4 0.2 0 -3.5 -3.3 -3.1 -2.9 -2.7 -2.5

As shown, the best fit line is log10 = 0.3792 log10 + 2.0518 Therefore, B = 102.0518 = 112.67 and m = 0.3792, and the power model is y = 112.67 x 0.3792 The model and the data can be plotted as
14 12 10 8 6 4 2 0 0 0.0005 0.001 0.0015

y = 112.67x 0.3792 R2 = 0.9997

0.002

0.0025

0.003

0.0035

20.51 Using linear regression gives

6 4 2 0 0 1 2 3 4 5 6 y = 0.688x + 2.79 R2 = 0.9773

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

41

Therefore, y = 2.79 and = 0.688. 20.52 A power fit can be developed as

12 10 8 6 4 2 0 0 20 40 60 80 100 120 140

y = 0.7374x0.5408 R2 = 0.9908

Therefore, = 0.7374 and n = 0.5408. 20.53 We fit a number of curves to this data and obtained the best fit with a second-order polynomial with zero intercept
5 4 3 2 1 0 0 0.004 0.008 0.012 0.016 0.02 0.024 y = 1628.1x2 + 140.05x R2 = 1

Therefore, the best-fit curve is u = 1628.1y 2 + 140.05 y We can differentiate this function du = 3256.2 y + 140.05 dy Therefore, the derivative at the surface is 140.05 and the shear stress can be computed as 1.8 105(140.05) = 0.002521 N/m2. 20.54 We can use transformations to linearize the model as

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

42 1 Ta

ln = ln D + B

Thus, we can plot the natural log of versus 1/Ta and use linear regression to determine the parameters. Here is the data showing the transformations.
T 0 5 10 20 30 40

1.787 1.519 1.307 1.002 0.7975 0.6529

Ta 273.15 278.15 283.15 293.15 303.15 313.15

1/Ta 0.003661 0.003595 0.003532 0.003411 0.003299 0.003193

ln 0.580538 0.418052 0.267734 0.001998 -0.22627 -0.42633

Here is the fit:

0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 0.0031

y = 2150.8x - 7.3143 R2 = 0.998

0.0032

0.0033

0.0034

0.0035

0.0036

0.0037

Thus, the parameters are estimated as D = e7.3143 = 6.65941 104 and B = 2150.8, and the Andrade equation is

= 6.65941 10 4 e 2150.8 / Ta
This equation can be plotted along with the data
2 1.5 1 0.5 0 0 10 20 30 40

Note that this model can also be fit with nonlinear regression. If this is done, the result is
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

43

= 5.39872 10 4 e 2210.66 / Ta
Although it is difficult to discern graphically, this fit is slightly superior (r2 = 0.99816) to that obtained with the transformed model (r2 = 0.99757). 20.55 Hydrogen: Fifth-order polynomial regression provides a good fit to the data,
15.2 15 14.8 14.6 14.4 14.2 14 200 300 400 500 600 700 800 900 1000 y = 4.3705E-14x5 - 1.5442E-10x4 + 2.1410E-07x3 - 1.4394E-04x2 + 4.7032E-02x + 8.5079E+00 R2 = 9.9956E-01

Carbon dioxide: Third-order polynomial regression provides a good fit to the data,
1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 200 300 400 500 600 700 800 900 1000 y = 4.1778E-10x3 - 1.3278E-06x2 + 1.7013E-03x + 4.4317E-01 R2 = 9.9998E-01

Oxygen: Fifth-order polynomial regression provides a good fit to the data,

1.1 1.05 1 0.95 0.9 200 300 400 500 600 700 800 900 1000 y = -1.3483E-15x + 5.1360E-12x - 7.7364E-09x + 5.5693E06x - 1.5969E-03x + 1.0663E+00 R2 = 1.0000E+00
2 5 4 3

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

44 Nitrogen: Third-order polynomial regression provides a good fit to the data,

1.2 1.16 1.12 1.08 1.04 1 200 300 400 500 600 700 800 900 1000 y = -3.0895E-10x + 7.4268E-07x - 3.5269E-04x + 1.0860E+00 R = 9.9994E-01
2 3 2

20.56 This part of the problem is well-suited for Newton interpolation. First, order the points so that they are as close to and as centered about the unknown as possible, for x = 4. y0 = 4 y1 = 2 y2 = 6 y3 = 0 y4 = 8 f(y0) = 38.43 f(y1) = 53.5 f(y2) = 30.39 f(y3) = 80 f(y4) = 30

The results of applying Newtons polynomial at y = 3.2 are

Order 0 1 2 3 4 f(y) 38.43 44.458 43.6144 43.368 43.48045 Error 6.028 -0.8436 -0.2464 0.112448

The minimum error occurs for the third-order version so we conclude that the interpolation is 43.368. (b) This is an example of two-dimensional interpolation. One way to approach it is to use cubic interpolation along the y dimension for values at specific values of x that bracket the unknown. For example, we can utilize the following points at x = 2. y0 = 0 y1 = 2 y2 = 4 y3 = 6 f(y0) = 90 f(y1) = 64.49 f(y2) = 48.9 f(y3) = 38.78

T(x = 2, y = 2.7) = 58.13288438 All the values can be tabulated as T(x = 2, y = 2.7) = 58.13288438
PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

45 T(x = 4, y = 2.7) = 47.1505625 T(x = 6, y = 2.7) = 42.74770188 T(x = 8, y = 2.7) = 46.5 These values can then be used to interpolate at x = 4.3 to yield T(x = 4.3, y = 2.7) = 46.03218664 Note that some software packages allow you to perform such multi-dimensional interpolations very efficiently. For example, MATLAB has a function interp2 that provides numerous options for how the interpolation is implemented. Here is an example of how it can be implemented using linear interpolation,
>> 85 70 55 40 >> >> >> T = Z=[100 90 80 70 60; 64.49 53.5 48.15 50; 48.9 38.43 35.03 40; 38.78 30.39 27.07 30; 35 30 25 20]; X=[0 2 4 6 8]; Y=[0 2 4 6 8]; T=interp2(X,Y,Z,4.3,2.7) 47.5254

It can also perform the same interpolation but using bicubic interpolation,
>> T=interp2(X,Y,Z,4.3,2.7,'cubic') T = 46.0062

Finally, the interpolation can be implemented using splines,

>> T=interp2(X,Y,Z,4.3,2.7,'spline') T = 46.1507

20.57 This problem was solved using an Excel spreadsheet and TrendLine. Linear regression gives
0.6 0.4 0.2 0 0 2 4 6 8 10 y = 0.0455x + 0.107 R2 = 0.9986

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.

46

Forcing a zero intercept yields

0.8 0.6 0.4 0.2 0 0 2 4 6 8 10 y = 0.061x R2 = 0.841

One alternative that would force a zero intercept is a power fit

0.8 0.6 0.4 0.2 0 0 2 4 6 8 10 y = 0.1821x0.4087 R2 = 0.8992

However, this seems to represent a poor compromise since it misses the linear trend in the data. An alternative approach would to assume that the physically-unrealistic non-zero intercept is an artifact of the measurement method. Therefore, if the linear slope is valid, we might try y = 0.0455x.

PROPRIETARY MATERIAL. The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed, reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a student using this Manual, you are using it without permission.