Scribd
Upload
66 views

Lab #2

This document appears to be a lab report for an Introduction to Matlab course. It includes 6 parts that cover topics like curve fitting, loading and analyzing data, differentiation and integration. For part 1, the student performed a linear curve fit on pressure transducer calibration data. For part 3, the student loaded data from a file to create a histogram and probability density graph. The student also calculated areas under curves to evaluate integration methods in Matlab versus by hand. Figures are included throughout to illustrate results.

Uploaded by

Pat O'Brien
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
66 views

Lab #2

This document appears to be a lab report for an Introduction to Matlab course. It includes 6 parts that cover topics like curve fitting, loading and analyzing data, differentiation and integration. For part 1, the student performed a linear curve fit on pressure transducer calibration data. For part 3, the student loaded data from a file to create a histogram and probability density graph. The student also calculated areas under curves to evaluate integration methods in Matlab versus by hand. Figures are included throughout to illustrate results.

Uploaded by

Pat O'Brien
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Course Number and Name:

ME 646: Experimental Measurement and Data Analysis



Semester and Year:
Spring, 2012
Name of Lab Instructor:
Ronald Goulet
Lab Section and Meeting Time:
Section 4: Thursday 2-5PM

Report Type:
Individual internal

Title of Experiment:
Introduction to Matlab
Date Experiment Performed:
Thursday, February 9
th
2012
Date Report Submitted:
Wednesday, February 15
th
2012
Names of Group Members:
Pat OBrien
Grader's Comments:


Grade:





Create PDF files without this message by purchasing novaPDF printer (http://www.novapdf.com)
Part II: Curve Fitting

Figure 1 illustrates calibration data from a Kistler 6125A piezoelectric pressure transducer. The
blue boxes were the samples taken while the line is a linear fit to the calibration data.

QII.1: The sensitivity of the transducer can be found by measuring the slope of the line.

y = mx + b, (1)

Using Eq. 1 wee found the sensitivity to be .002

QII.2: Only after we zoomed in a great amount on the graph could we notice a change due to
hysteresis, therefore we decided it was valid to neglect hysteresis.

QII.3:
u = yku
c
(4)

Using Eq. 4 we obtain a value of (+)(-) .016kPa for a measurement of uncertainty.
























Figure 1 Calibration data from a 6125A piezoelectric pressure transducer
Create PDF files without this message by purchasing novaPDF printer (http://www.novapdf.com)
Part III: Load data from a file and Create a Histogram

Figure 2 shows the histogram produced from the data in the me646_data.txt file using 50 bins.

Figure 3 shows the (X,Y) plot formed from the histogram shown in Fig. 2.

QIII.1 Fig. 3 shows a smoother data set compared to the histogram in Fig.

QIII.2 Fig. 4 resembles a normal distribution or a Gaussian probability density function.

QIII.3
a) We calculated our area using Fig. 4 and the equation below.
[
1
2
b = Arco (2)
After using the above equation and the graph shown in Fig 4, we came up with an area of
.15 meaning 15% chance the data point will fall between 1 and 2 and the graph.
b) Do not have table 4.3
c) Do not have table 4.3

QIII.4 Using the same method as we did in question 3, we used Eq. 2 to solve for the area under
the whole curve. We came up with an area of .95 or 95% chance that the data point will fall
under that curve. The number theoretically should be 1 or 100% so our method was fairly
accurate.






Figure 2 The Histogram produced from our me646_data.txt file that
we loaded into Matlab
Figure 3 The (X,Y) graph that was produced from our Histogram
Figure 4 The yPrime graph from our data set
Create PDF files without this message by purchasing novaPDF printer (http://www.novapdf.com)
Part VI: Differentiation and Integration
Figure 5 shows the sine and cosine function, as well as the derivative of the sine function.
Figure 6 shows the sine function, but with less sampling points. The graph illustrates what
happens when liaising occurs.
QVI.1: Using the blue curve seen in Fig. 5, we calculated the area isolated by the curve by using
Eq. 2 on the previous page. We found an area of 6.28.
QVI.2: As seen in Fig. 5, the green line and the red line are overlapped. This is because the
derivative of sine (the red circles) and the cosine function equal the same value.
QVI.3: Deri.ms gradient matches that of Matlabs gradient.
QVI.4: Deri.ms area calculation was -1.838e^-15 while Matlabs area calculation was
-3.984e^-17. Although the numbers are different, they are both extremely close to zero.
QVI.5:
si n( x) Jx = cos( x)
2n
-2n
(3)
Evaluating Eq. 3, from -2 to 2 = 1-1 = 0. Calculating the area by hand gives us a perfect zero
output. It is more accurate than Matlabs because Matlab has minute interfering inputs.
QVI.6: When you decrease the sampling rate to a small enough number you begin to see
aliasing. This produces an inaccurate graph as seen in Fig. 6.
Figure 5 Blue line represents the Sine function. The green
line represents the cosine function. The red circles
represent the derivative of the sine function.
Figure 6 Sine function after the n value, or the sampling
rate, has been decreased.
Create PDF files without this message by purchasing novaPDF printer (http://www.novapdf.com)
Introduction to JMP

Part VII: Opening Data Files and Creating Fit Curve

QVII.1: JMP produces the same graph, however, JMP provides us with data points on the graph.

Part VIII: Loading Data from Excel and Distribution Analysis

QVIII.1: There are very few points outside of the 95% accuracy range because the data was
recorded with a very accurate device. There is very little fluctuation from the median with
accurate instruments so the 95% accuracy range leave a fair amount of room for the data points
to stray without leaving the accuracy range.

QVIII.2: All of the graphs represent a normal, or Gaussian distribution

QVIII.3: Increasing the sampling rate increases the accuracy of the probability density functions
graph.
Create PDF files without this message by purchasing novaPDF printer (http://www.novapdf.com)

You might also like