Research on information fusion method for heat model and weather model based on HOGA-SVM.

Modeling the heat released from a burning wildfire is essential to support an accurate representation of the heat flux of wildfire. At present, the commonly used heat model is combustion heat modeling in the wildfire spread model of DEVS-FIRE. DEVS-FIRE employs Rothermel's model to calculate the rate of fire spread, but it is poorly suitable for the actual situation in China. Forest fire spreading is a dynamic and complex system. In the process of spreading, the heat released by forest fire will affect the weather conditions in the current fire area, and the weather conditions such as temperature, humidity, wind speed and wind direction are the important factors affecting the spread of forest fire in turn. Therefore, the interaction between heat and weather conditions must be considered in order to accurately predict the spread of forest fires. This paper proposes heat model based on Zhengfei Wang's wildfire spread model. The heat model was used to fuse with the weather model, and the information was fused based on the hybrid orthogonal genetic algorithm (HOGA) and support vector machine (SVM). The fused data was input into DEVS-Fire to simulate the spread of wildfire. Experiment results demonstrate that this method improves the precision of wildfire spread. [ABSTRACT FROM AUTHOR]