Scribd
Upload
48 views

University of The Witwatersrand, Johannesburg School of Mechanical, Industrial & Aeronautical Engineering

This document provides instructions for a MatLab project that involves analyzing historical data on valve seat diameters from engine blocks. Students are asked to create a program that imports diameter measurement data from a .csv file, normalizes the data, performs linear regression to determine an optimal tool replacement frequency, and outputs the results. The program must include a graphical user interface and allow the user to iteratively test different data sets until a satisfactory replacement frequency is determined. Consultation for the project is available on specified dates in room NWE F20.

Uploaded by

Ali Gh
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
48 views

University of The Witwatersrand, Johannesburg School of Mechanical, Industrial & Aeronautical Engineering

This document provides instructions for a MatLab project that involves analyzing historical data on valve seat diameters from engine blocks. Students are asked to create a program that imports diameter measurement data from a .csv file, normalizes the data, performs linear regression to determine an optimal tool replacement frequency, and outputs the results. The program must include a graphical user interface and allow the user to iteratively test different data sets until a satisfactory replacement frequency is determined. Consultation for the project is available on specified dates in room NWE F20.

Uploaded by

Ali Gh
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

1

University of the Witwatersrand, Johannesburg


School of Mechanical, Industrial & Aeronautical Engineering
MECN2003: Software Development
MatLab Project
Due Date: 2010-03-23 17:00
Problem Description
When machining valve seats in cast engine blocks, an open loop tool replacement policy is
used, meaning that the machining tool is replaced after a fixed number of valve seats have
been machined. The setting of the machine may mean that some valve seats machined at the
beginning of the tool life are machined too large (i.e. larger than the maximum diameter allowed
for proper sealing) while the wear of the tool means that some of the valve seats machined near
the end of the tool life may be too small (i.e. smaller than the minimum diameter). In cases
where the valve seat is too small the blocks in question are sent for re-machining to bring them
into spec (i.e. within the specified limits of maximum and minimum valve seat diameter). In
cases where the valve seat is too large the blocks are scrapped.
Under the current manufacturing model the machine tool is replaced every 30 valve seats. An
investigation is to be made of current data to determine if the frequency of tool changes could
be changed to reduce the number of blocks sent both for re-machining or scrap. A program
is required which will ask the user for the nominal valve seat diameter required as well as
the minimum and maximum acceptable diameters. The program must then ask the user for a
Comma-Separated Value (.csv) file of historical data of valve seat diameters measured under the
current open-loop replacement model. These data sets must be zeroed relative to the minimum
permissible valve seat diameter (i.e. how much each valve seat is bigger than the minimum
diameter limit) and sorted such that all sets are plotted on the same set of axes reflecting the
adjusted diameter as a function of the cycle number (i.e. valve seat machined by a particular
tool). This condensed data set, which may be comprised of any number of tool cycle data sets,
must be plotted and displayed to the user along with a linear regression suggesting the new
cycle frequency.
If the user feels that the presented regression is unacceptable for whatever reason (too many
over-sized valve seats, new replacement frequency too short or too long...) the user must be
asked to select a new input .csv file and to repeat this process until they are satisfied that the new
replacement frequency is reasonable. Once the user is satisfied with the presented regression
is acceptable the program must report the overall linear polynomial, the root thereof and the
resulting new replacement frequency in an automatically-generated text file where the name is
based on the input file name e.g. if the input .csv file was InputData.csv then the output file
must be named InputDataFit.txt and be saved in the same directory as the input file.
All user interface must be via GUIs. Bonus marks, up to a maximum of 10%, are allowed for
features that make the program more generically applicable, flexible or decidedly robust.

2
Supplied Material
Sample file (Sample.csv) of machining data based on a nominal valve seat diameter of
24mm, a minimum valve seat diameter of 23.980mm and maximum valve seat diameter
of 24.020mm. The files to be analysed by this program will be in a two-column format
listing valve seat number and measured diameter.
Deliverables
Each student is to submit only a single m-file code named ValveSeat[StudentNumber].m
e.g. For John Smith, student number 0701234A, the code would be named ValveSeat0701234A.m
(If file extensions are not shown on the computer used for development the file must be
named ValveSeat0701234A otherwise the code will not run. Please clarify whether this
is the case with one of the tutors or Mr Paton)
Each code must contain an opening comment with the student name and student number
as well as the date of finalization of the code
Codes are to be submitted via the Ignite system under the MatLab Project item. Any code
received not following the outlined naming convention and submission instructions
will not be marked and a mark of 0 will be awarded for the project. Each student
may submit only 1 code.
Consultation Process
Consultation for the project will be at the following times. e-mail correspondence for consultative purposes will be limited and, in general, any unsolicited correspondence will result
in penalties against the project mark.
Table 1: Consultation Times for MECN2003 MatLab Project
Date
Time
Venue
2010-03-10* 14:00 - 17:00 PC Pool
2010-03-15
14:00 - 17:00 NWE F20
2010-03-17
14:00 - 17:00 NWE F20
2010-03-19
14:00 - 17:00 NWE F20
A roster will be posted outside room SWE101 where students can book 10-minute consultation
slots on the relevant days. Failure to arrive promptly for a consultation will result in the time
being given to the next waiting student.
* Subject to the completion of MatLab Laboratory 2

You might also like