Forest fire risk prediction model based on machine learning

doi:10.16265/j.cnki.issn1003-3033.2022.09.2729

Abstract

Abstract:

In order to reduce the damage caused by forest fires, the forest fire risk was modeled and predicted through a literature review. This paper summarizes the research status of forest fire risk prediction based on the machine learning method, and analyzes and expounds on the selection of forest fire impact factors, the selection of appropriate fire risk prediction models and the model test methods. The results show that the main influencing factors of forest fire risk include the characteristics of combustibles, meteorological factors, terrain, human activities, etc. In the forest fire risk prediction model, the back propagation(BP) neural network method needs to be improved and applied. The SVM method requires high data, the random forest(RF) method has strong universality and high accuracy, and the deep learning method has less research, but the accuracy is high. The commonly used test methods of the model are accuracy, receiver operating characteristic(ROC) curve and area under curve(AUC) value.

Key words: machine learning, forest fire risk prediction, forest fire, meteorological factors, support vector machine(SVM)

ZHU Xin, LI Jianwei, GUO Wei, BI Sheng, WU Yuefei. Forest fire risk prediction model based on machine learning[J]. China Safety Science Journal, 2022, 32(9): 152-157.

Figures/Tables 2

References 30

[1]	龙腾腾, 殷继艳, 欧朝蓉, 等. 云南省森林火灾风险综合评价及空间格局研究[J]. 中国安全科学学报, 2021, 31(9):167-173. doi: 10.16265/j.cnki.issn1003-3033.2021.09.023
	LONG Tengteng, YIN Jiyan, OU Zhaorong, et al. Comprehensive assessment and spatial pattern study on forest fire risk in Yunnan Province[J]. China Safety Science Journal, 2021, 31(9):167-173. doi: 10.16265/j.cnki.issn1003-3033.2021.09.023
[2]	舒立福, 张小罗, 戴兴安, 等. 林火研究综述(Ⅱ):林火预测预报[J]. 世界林业研究, 2003(4):34-37.
[3]	田晓瑞, DOUGLAS J, 张有慧. 森林火险等级预报系统评述[J]. 世界林业研究, 2006(2):39-46.
[4]	曹姗姗. 小尺度森林火险等级预报模型研究[D]. 北京: 中国林业科学研究院, 2014.
	CAO Shanshan. Small-scale forest fire danger rating prediction model[D]. Beijing: Chinese Academy of Forestry, 2014.
[5]	GIGOVIC L, POURGHASEMI H, DROBNJAK S, et al. Testing a new ensemble model based on SVM and random forest in forest fire susceptibility assessment and its mapping in Serbia's tara national park[J]. Forests, 2019, 10(5):408-429. doi: 10.3390/f10050408
[6]	高超, 林红蕾, 胡海清, 等. 我国林火发生预测模型研究进展[J]. 应用生态学报, 2020, 31(9):3227-3240. doi: 10.13287/j.1001-9332.202009.014
	GAO Chao, LIN Honglei, HU Haiqing, et al. A review of models of forest fire occurrence prediction in China[J]. Chinese Journal of Applied Ecology, 2020, 31(9):3227-3240. doi: 10.13287/j.1001-9332.202009.014
[7]	王璠. 基于改进BP神经网络的林火预测研究[D]. 南昌: 江西科技师范大学, 2016.
	WANG Fan. Study on forest fire prediction based in improved BP neural network[D]. Nanchang: Jiangxi Science and Technology Normal University, 2016.
[8]	马文苑. 大尺度林火驱动因子及预测模型研究[D]. 北京: 北京林业大学, 2019.
	MA Wenyuan. Study on driving factors and forecast models of forest fire in large-scale areas[D]. Beijing: Beijing Forestry University, 2019.
[9]	蒋琴, 钟少波, 朱伟. 京津冀地区森林火灾综合风险评估[J]. 中国安全科学学报, 2020, 30(10):119-125. doi: 10.16265/j.cnki.issn 1003-3033.2020.10.017
	JIANG Qin, ZHONG Shaobo, ZHU Wei. Research on comprehensive risk assessment of forest fire in Beijing-Tianjin-Hebei region[J]. China Safety Science Journal, 2020, 30(10): 119-125. doi: 10.16265/j.cnki.issn 1003-3033.2020.10.017
[10]	汤静. 基于BP神经网络的森林火险预测系统[D]. 重庆: 重庆大学, 2002.
	TANG Jing. The foerst fire size class forecasting system on BP neural net[D]. Chongqing: Chongqing University, 2002.
[11]	胡超. 基于BP人工神经网络的区域森林火灾预测研究[D]. 舟山: 浙江海洋学院, 2015.
	HU Chao. The regional forest fire research based on BP neural network[D]. Zhoushan: Zhejiang Ocean University, 2015.
[12]	王亚琴. BPNN-DIOC网络优化理论及其森林火险预测研究[D]. 太原: 太原理工大学, 2018.
	WANG Yaqin. Reseach on BPNN-DIOC network optimization theory and its forest fire rest prediction[D]. Taiyuan: Taiyuan University of Technology, 2018.
[13]	丁世飞, 齐丙娟, 谭红艳. 支持向量机理论与算法研究综述[J]. 电子科技大学学报, 2011, 40(1):2-10.
	DING Shifei, QI Bingjuan, TAN Hongyan. An overview on theory and algorithm of support vector machines[J]. Journal of University of Electronic Science and Technology of China, 2011, 40(1):2-10.
[14]	黄玉霞, 许东蓓, 蒲肃. SVM方法在森林火险预测中的应用[J]. 林业科学, 2007(10):77-82.
	HUANG Yuxia, XU Dongbei, PU Su. Application of SVM method on the prediction of forest fire danger[J]. Scientia Silvae Sinicae, 2007(10):77-82.
[15]	李娴. LS-SVM模型选择及其在森林火灾预测中的应用[J]. 统计与决策, 2011(21):163-164.
[16]	许志卿, 苏喜友, 张颐. 基于支持向量机方法的森林火险预测研究[J]. 中国农学通报, 2012, 28(13):126-131.
	XU Zhiqing, SU Xiyou, ZHANG Yi. Forest fire prediction based on support vector machine[J]. Chinese Agricultural Science Bulletin, 2012, 28(13):126-131.
[17]	SINGH K R, NEETHU K P, MADHUREKAA K, et al. Parallel SVM model for forest fire prediction[J]. Soft Computing Letters, 2021, 3:DOI: 10.1016/J.SOCL.2021.100014. doi: 10.1016/J.SOCL.2021.100014
[18]	梁慧玲, 林玉蕊, 杨光, 等. 基于气象因子的随机森林算法在塔河地区林火预测中的应用[J]. 林业科学, 2016, 52(1):89-98.
	LIANG Huiling, LIN Yurui, YANG Guang, et al. Application of random forest algorithm on the forest fire prediction in Tahe area based on meteorological factors[J]. Scientia Silvae Sinicae, 2016, 52(1):89-98.
[19]	GUO Futao, WANG Guangyu, SU Zhangwen, et al. What drives forest fire in Fujian, China? evidence from logistic regression and random forests[J]. International Journal of Wildland Fire, 2016,25, 505-519.
[20]	潘登, 郁培义, 吴强. 基于气象因子的随机森林算法在湘中丘陵区林火预测中的应用[J]. 西北林学院学报, 2018, 33(3):169-177.
	PAN Deng, YU Peiyi, WU Qiang. Application of random forest algorithm on the forest fire prediction based on meteorological factors in the Hilly area,central Hunan peovince[J]. Journal of Northwest Forestry University, 2018, 33(3):169-177.
[21]	侯晓静, 明金科, 秦荣水, 等. 基于随机森林模型的交界域火灾风险分析[J]. 林业科学, 2019, 55(8):194-200.
	HOU Xiaojing, MING Jinke, QIN Rongshui, et al. Analysis of the fire risk in wildland-urban interface with random forest model[J]. Scientia Silvae Sinicae, 2019, 55(8):194-200.
[22]	马文苑, 冯仲科, 成竺欣, 等. 山西省林火驱动因子及分布格局研究[J]. 中南林业科技大学学报, 2020, 40(9):57-69.
	MA Wenyuan, FENG Zhongke, CHENG Zhuxin, et al. Study on driving factors and distribution pattern of forest fires in Shanxi province[J]. Journal of Central South University of Forestry & Technology, 2020, 40(9):57-69.
[23]	尹宝才, 王文通, 王立春. 深度学习研究综述[J]. 北京工业大学学报, 2015, 41(1):48-59.
	YIN Baocai, WANG Wentong, WANG Lichun. Review of deep learning[J]. Journal of Beijing University of Technology, 2015, 41(1):48-59.
[24]	南玉龙, 张慧春, 郑加强, 等. 深度学习在林业中的应用[J]. 世界林业研究, 2021, 34(5):87-90.
	NAN Yulong, ZHANG Huichun, ZHENG Jiaqiang, et al. Research progress of deep learning in forestry application[J]. World Forestry Research, 2021, 34(5):87-90.
[25]	张荣, 李伟平, 莫同. 深度学习研究综述[J]. 信息与控制, 2018, 47(4):385-397,410.
	ZHANG Rong, LI Weiping, MO Tong. Review of deep learning[J]. Information and Control, 2018, 47(4):385-397,410.
[26]	孙立研, 刘美玲, 周礼祥, 等. 基于气象因子深度学习的森林火灾预测方法[J]. 林业工程学报, 2019, 4(3):132-136.
	SUN Liyan, LIU Meiling, ZHOU Lixiang, et al. Research on forest fire prediction method based on deep learning[J]. Journal of Forestry Engineering, 2019, 4(3):132-136.
[27]	ZHANG Guoli, WANG Ming, LLU Kai. Forest fire susceptibility modeling using a convolutional neural network for Yunnan province of China[J]. International Journal of Disaster Risk Science, 2019, 10(3):386-403. doi: 10.1007/s13753-019-00233-1
[28]	张萌. 基于深度学习的森林火险预测研究[D]. 北京: 北京林业大学, 2020.
	ZHANG Meng. Forest fire risk prediction based on deeplearning[D]. Beijing: Beijing Forestry University, 2020.
[29]	LI Yudong, FENG Zhongke, CHEN Shilin, et al. Application of the artificial neural network and support vector machines in forest fire prediction in the Guangxi autonomous region, China[J]. Discrete Dynamics in Nature and Society, 2020, 2020,1-14.
[30]	MOHAJANE M, COSTACHE R, KARIMI F, et al. Application of remote sensing and machine learning algorithms for forest fire mapping in a Mediterranean area[J]. Ecological Indicators, 2021, 129:107869-107 886.

影响因素	自变量	单位
可燃物特征	可燃物类型	类
	可燃物含水量	kg/m²
	植被覆盖率	%
	腐殖质湿度	kg/m²
	NDVI	—
地形	坡度	(°)
	坡向	(°)
	高程	m
	海拔	m
气象	温度	℃
	相对湿度	%
	风速	m/s
	降水量	mm/h
	干旱时间	d
人类活动	与道路网距离	m
	与居民点距离	m
	人口密度	人/km²

预测模型		优点		缺点		适用性
BP神经网络		具备神经网络的普遍优点,非线性映射能力和抗干扰能力		学习时间过长,容易陷入局部最小值,评价结果不确定		不适应与少量林火数据,适合气象因子的预测
SVM		可将输入向量映射到高维空间,解决最优分类问题		当林火因子数增加时,计算速度降低,难以多分类和大规模数据进行处理		适合火险因子较少的样本
RF	无需进行多重共线性检验,具有独立的模型检验方法,能避免过拟合和欠拟合		无法得知内部如何运行,没有函数表达式		普遍适用
深度学习	优秀的图像处理能力,可以考虑相邻空间的相关性		需要大量的训练样本,计算效率低下		适合矢量与栅格图像数据的处理