This preview shows pages 1–2. Sign up to view the full content.
16.881 – Robust System Design
Solution to Quiz #7
Design of Dynamic Systems
You are designing a light truck steering system. Customer surveys indicate that drivers
prefer a consistent response of the steering system under a wide range of driving
conditions (speeds, road surfaces, weather conditions, etc). The following data were
collected to evaluate a baseline design:
Turning radius (feet)
Steering wheel
angle (degrees)
Noise condition
level 1
Noise condition
level 2
Noise condition
level 3
5
330
336
333
45
72
74
77
90
37
38
36
1)
What problems might you run into if you performed a continuouscontinuous
dynamic signal to noise analysis using steering wheel angle as the signal and turning
radius as the response? What would you choose as the signal factor and response of
the system to make a continuouscontinuous dynamic signal to noise analysis
feasible? Sketch a graph (or a few graphs) to support your choice.
A sketch of the data shows that there is very little variance due to the induced noise as
compared to the variation caused by the signal factor. In other words, this system seems
to be very repeatable. However, if the turning radius is taken as the response, the
relationship is highly nonlinear and has a nonzero y intercept.
If you use the Taguchi definition of
β
which is essentially a regression with a zero y
intercept, you get a very poor fit to the data. Error variance will be extremely high.
You can get a better fit by allowing for a nonzero y intercept. However, the clear non
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
 Fall '03
 lDavidMiller

Click to edit the document details