Two important applications of the fiber Fabry-Perot Interferometer (FFPI) sensor
are investigated: (1) an optical binary switch for aerospace application, and (2) an FFPI
weigh-in-motion sensor for measuring the weight of trucks traveling down a highway.
In the fiber optical switch, the FFPI sensor is bonded to a copper cantilever to
sense the elongation of cavity length induced by force applied to the end of the
cantilever via a pushed button. Light from a superluminescent diode light source passes
through a scanned Michelson interferometer and is reflected from a sensing FFPI and a
reference FFPI to produce a fringe pattern. A secondary interferometer uses a
distributed feedback laser light source to compensate for irregularities in the mechanical
scanning rate of the moving stage to achieve precision measurement of the optical path
difference.
The system is calibrated by applying known weights to the cantilever. The
elongation measured by the FFPI sensor shows excellent linearity as a function of the force applied, and little hysteresis was observed. By comparing the measured force to a
threshold, the system produces a binary signal that indicates the state of the pilotactuated
system; i. e., whether or not the button has been pushed.
In FFPI weigh-in-motion sensors system, the FFPI sensors are installed in metal
bars so that they will experience the strain induced by applied loads and are connected to
the Signal Conditioning Unit (SCU). The SCU determines the induced phase shift in the
FFPI and produces voltage outputs proportional to the phase shifts.
Laboratory Material Testing System tests show that the fiber optic sensor
response is a fairly linear function of the axial displacement. In highway tests the FFPI
sensors showed strong responses and consistently reproduced the expected
characteristics of truck wheel crossings. A falling weight deflectometer was used to
calibrate the sensor response and predict unknown loads. All sensors in steel bars and
aluminum bars showed excellent repeatability and accurate predictions, with an average
relative percentage error within 2%. The study on sensor response variation with applied
load positions shows a bell shaped distribution. Truck tests on the road sensors indicate
that the repeatability of wheel crossings at similar position is good. The sensor can
accurately measure axle spacing, speed, and truck class.
| Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/ETD-TAMU-1019 |
| Date | 15 May 2009 |
| Creators | Xie, Zhaoxia |
| Contributors | Eknoyan, Ohannes, Taylor, Henry |
| Source Sets | Texas A and M University |
| Language | en_US |
| Detected Language | English |
| Type | Book, Thesis, Electronic Dissertation, text |
| Format | electronic, application/pdf, born digital |
Page generated in 0.0014 seconds