Show simple item record

dc.contributor.advisorAl-Anbuky, Adnan
dc.contributor.authorKoppula, Lakshmi Bhargavi
dc.date.accessioned2012-10-23T02:16:23Z
dc.date.available2012-10-23T02:16:23Z
dc.date.copyright2011
dc.date.created2012
dc.date.issued2012-10-23
dc.identifier.urihttp://hdl.handle.net/10292/4661
dc.description.abstractThis thesis investigates the topic of “Wildfire hazard prediction” through conducting an in-depth study on fuzzy prediction methods and geographically collected weather data. The study explores the impact of various environmental factors leading to Wildfire. These factors associated with Wildfire are extracted from analyzing the past raw weather data and using McArthur’s Fire Danger Index formulations. The indices calculated through the formula and the generated synthetic data are used to train a Fuzzy system developed in Matlab software. The trained Fuzzy system is then tested with a raw set of historical real weather data originated from National Rural Fire Authority (NRFA) and National Institute of Water and Atmospheric Research (NIWA) to analyze the accuracy of the system developed. Finally, the predicted results of the Fuzzy system are examined and compared with that calculated using the formula, including the error percentiles between the two. Impacts of the input weather factors are also plotted in relation to the Fire Danger Indices under various conditions to understand their sensitivity towards the final prediction.en_NZ
dc.language.isoenen_NZ
dc.publisherAuckland University of Technology
dc.subjectPredictionen_NZ
dc.subjectFuzzyen_NZ
dc.subjectMATLABen_NZ
dc.subjectWildfireen_NZ
dc.subjectBushfireen_NZ
dc.titleWildfire hazard prediction: a fuzzy model for Sensor Embedded Intelligenceen_NZ
dc.typeThesis
thesis.degree.grantorAuckland University of Technology
thesis.degree.levelMasters Theses
thesis.degree.nameMaster of Engineeringen_NZ
thesis.degree.discipline
dc.rights.accessrightsOpenAccess
dc.date.updated2012-10-22T23:50:49Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record