In order to better prevent fire and adapt to the change of the probability of extreme fire events in the next ten years, we establish an improved grey neural network prediction model. According to the collected data of climate characteristics, precipitation, temperature, humidity, wind speed and air pressure in southeastern Australia, they are quantified and processed as learning samples, and the improved grey neural network is used to predict the scale of wildfires in southeastern Australia. Through sensitivity analysis, the robustness of the model is guaranteed.

Wei Li, Yuchi Li, Xinshuo Du, Shumin Liu, Xin He.Research on Field Fire Prediction Based on Improved Grey Correlation Prediction Based on BP Neural Network. Academic Journal of Computing & Information Science (2022), Vol. 5, Issue 6: 14-18. https://doi.org/10.25236/AJCIS.2022.050603.

[1] Sun Shouqin. Why did Australia’s forest fires get out of control this time? [J]. Fire Fighting (Electronic Edition), 2020. 

[2] Wang Miao, Wang Pengya, Chen Xin. U.S. carbon emission prediction based on gray neural network model J]. Science and Technology Information, 2019. 

[3] Yang Lin, Kong Feng. Hybrid artificial bee colony algorithm embedded with particle swarm optimization algorithm [J]. Automatic Instrumentation, 2013. 

[4] Yin Jianxia, Meng Hongyun. Artificial bee colony algorithm with chaotic differential evolution search [J].Computer Engineering and Applications, 2011. 

[5] Zhang Dongnan, Shu Zhongjun, Chen Qingquan, Xu Yan. Application research of improved grey neural network in fire prediction J]. Chinese Safety Science Journal, 2012. 

[6] Wang Wenye, Liu Jing, Jia Nan. Forest fire prediction analysis based on neural network[J]. Journal of Armed Police Academy, 2020. 

[7] Li Chengxin, Zhao Liangchen, Li Dongyun, Ma Jiahui. Australian forest firespreadprediction model based on CA system and application of emergency supplies dispatching [J]. Forestry Inventory and Planning, 2020.