Input Parameter Calibration in Forest Fire Spread Prediction: Taking the Intelligent Way
Abstract
Imprecision and uncertainty in the large number of input parameters are serious problems in forest fire behaviour modelling. To obtain more reliable forecasts, fast and efficient computational input parameter estimation and calibration mechanisms should be integrated. These have to respect hard real-time constraints of simulations to prevent tragedy. We propose an Evolutionary Intelligent System (EIS) for parameter calibration. Depending on disaster size, required parameter precision, and available computing resources, the hybridisation of an evolutionary algorithm (EA) with an intelligent paradigm (IP) can be configured. Experiments show that EIS generates comparable estimations to standard evolutionary calibration approaches, clearly outperforming the latter in runtime.
Cite
Text
Wendt and Cortés. "Input Parameter Calibration in Forest Fire Spread Prediction: Taking the Intelligent Way." International Joint Conference on Artificial Intelligence, 2011. doi:10.5591/978-1-57735-516-8/IJCAI11-503Markdown
[Wendt and Cortés. "Input Parameter Calibration in Forest Fire Spread Prediction: Taking the Intelligent Way." International Joint Conference on Artificial Intelligence, 2011.](https://mlanthology.org/ijcai/2011/wendt2011ijcai-input/) doi:10.5591/978-1-57735-516-8/IJCAI11-503BibTeX
@inproceedings{wendt2011ijcai-input,
title = {{Input Parameter Calibration in Forest Fire Spread Prediction: Taking the Intelligent Way}},
author = {Wendt, Kerstin and Cortés, Ana},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2011},
pages = {2862-2863},
doi = {10.5591/978-1-57735-516-8/IJCAI11-503},
url = {https://mlanthology.org/ijcai/2011/wendt2011ijcai-input/}
}