Abstract
Wildfire spread in living vegetation such as chaparral in California and eucalyptus forests in Australia often causes significant damage to infrastructure and ecosystems. A physically based empirical model to predict fire spread rate is used in the United States to assist in a variety of fire management operations. The spread model does not adequately describe the chemical processes of combustion in live fuels.
Prior to describing and modeling the chemical processes of combustion in wildland fuels using computational fluid dynamics, we are investigating a technique to non-intrusively measure flame gas velocities using thermal imagery. By tracing "hot" pixels through successive digital images, we estimate velocity field using gradient-based algorithms. We also explore techniques established in digital particle image velocimetry (DPIV) to estimate fluid velocities. The images are acquired by a thermal camera with uncooled microbolometer 320x240 pixel focal plane array in the 7.5 - 13 um spectral range. We estimated fluid velocities in flames spreading above isopropyl alcohol and shredded aspen wood (excelsior). Results from excelsior fires are presented. Finally, results obtained from computational modeling were used to validate the velocity field estimated from the gradient-based algorithm. Preliminary results using a DPIV-based algorithm appear promising.
Original language | American English |
---|---|
Journal | Default journal |
State | Published - Mar 1 2002 |
Keywords
- IR-Based estimation
- chaparral
- DPIV
Disciplines
- Other Engineering