Prediction of Shrubbery Fire in Central Yunnan Province Based on the Adaptive NeuroFuzzy Inference System

doi:10.13466/j.cnki.lyzygl.2022.02.015

Abstract

Abstract:

The original vegetation in central Yunnan has been seriously damaged,leading to the continuous growth of flammable shrubbery. With the intensification of global warming,forest fires occur frequently,mainly in shrubs. Therefore,the prediction of forest fires is of great importance to the protection of forest resources in central Yunnan. Based on the data of shrubbery fire and its corresponding meteorological data from 1999 to 2019 in central Yunnan,China.Adaptive Neuro Fuzzy Inference System (ANFIS) and Logistic Regression (LR) were selected,using MATLAB,SPSS 25 and other software to establish a prediction model of forest fire in central Yunnan based on meteorological factors(70% of the data),then the remaining 30% of the data was tested.The results showed that nine meteorological factors were formed into three principal components by principal component analysis. As the input factors of ANFIS model,3 principal components could explain 77.663% of the information of 9 meteorological factors. According to the multicollinearity test,the VIF of the LR model was less than 10,and the precipitation at 24 o'clock,the average 2-minute wind speed,the average daily relative humidity and the daily minimum relative humidity could be obtained,then they were inputted into the model as independent variables of the LR model. According to the screening results of meteorological factors of the two models,the main influencing factors of shrubbery fire in central Yunnan were temperature,wind speed and humidity. By comparing the results of ANFIS and LR model,the accuracy of ANFIS training set was 12% higher than that of LR model,and the accuracy of test set was 10% higher than that of LR model. The ANFIS training set value was 0.96,the test set AUC value was 0.88,and the LR model AUC values were 0.875 and 0.816,respectively. The results showed that the ANFIS model had better adaptability to the model of forest fire in central Yunnan. Results of the research can provide some scientific basis for shrubbery fire prediction in central Yunnan.

Key words: meteorological factors, logistic regression, Adaptive Neuro Fuzzy Inferences System, shrubbery fire, central Yunnan Province

CLC Number:

S711
S762.2

WEI Jianheng, ZHAO Heng, GAO Zhongliang, WANG Hechenyang, MA Zenan, WANG Qiuhua, SHU lifu, YANG Hongmei. Prediction of Shrubbery Fire in Central Yunnan Province Based on the Adaptive NeuroFuzzy Inference System[J]. FOREST RESOURCES WANAGEMENT, 2022, (2): 109-116.

Figures/Tables 8

Fig.1

Tab.1

Tab.2

Fig.2

Tab.3

Tab.4

Tab.5

Fig.3

References 25

[1]	魏书精, 罗斯生, 罗碧珍, 等. 气候变化背景下森林火灾发生规律研究[J]. 林业与环境科学, 2020, 36(2):133-143.
[2]	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.
[3]	陈岱. 基于Logistic回归模型的大兴安岭林火预测研究[J]. 林业资源管理, 2019(1):116-122.
[4]	蔡奇均, 曾爱聪, 苏漳文, 等. 基于Logistic回归模型的浙江省林火发生驱动因子分析[J]. 西北农林科技大学学报:自然科学版, 2020, 48(2):102-109.
[5]	苏漳文, 刘爱琴, 梁慧玲, 等. 基于气象因子的福建省森林火险预测模型[J]. 森林与环境学报, 2015, 35(4):370-376.
[6]	苏漳文, 曾爱聪, 蔡奇均, 等. 基于Gompit回归模型的大兴安岭林火预测模型及驱动因子研究[J]. 林业工程学报, 2019, 4(4):135-142.
[7]	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
[8]	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.
[9]	Jang J S R. ANFIS:adaptive-network-based fuzzy inference system[J]. IEEE Transactions on Systems,Man,and Cybernetics, 1993, 23(3):665-685 doi: 10.1109/21.256541
[10]	黄庆键, 欧周, 林佳亮. 关于空间电力负荷预测方法综述与展望[J]. 自动化应用, 2017(2):79-81.
[11]	林涛, 刘航鹏, 赵参参, 等. 基于SSA-PSO-ANFIS的短期风速预测研究[J]. 太阳能学报, 2021, 42(3):128-134.
[12]	尹宝全, 曹闪闪, 傅泽田, 等. 水产养殖水质检测与控制技术研究进展分析[J]. 农业机械学报, 2019, 50(02):1-13.
[13]	Jaafari A, Termeh S, Bui D T. Genetic and firefly metaheuristic algorithms for an optimized neuro-fuzzy prediction modeling of wildfire probability[J]. Journal of Environmental Management, 2019, 243:358-369. doi: 10.1016/j.jenvman.2019.04.117
[14]	王秋华, 赵凤君, 肖慧娟, 等. 滇中安宁地区火烧迹地光叶石栎灌木林的火行为模拟[J]. 沈阳农业大学学报, 2015, 46(1):105-109.
[15]	李世友, 王少名, 年有春, 等. 落叶灌木南烛在防火戒严期的燃烧性动态[J]. 江西农业大学学报, 2008(5):845-849.
[16]	林向东. 昆明市石林县“2·7”森林火灾扑火战例评析[J]. 森林防火, 2012(1):38-41.
[17]	吉哲鹏. 云南安宁“5·21”森林火灾明火全部扑灭[EB/OL].(2014-05-26)[2022-01-07]. http://society.workercn.cn/2/201405/26/140526143837598.shtml .
[18]	张瑞坤, 昆明市禄劝县突发森林火灾 800余人紧急扑救[EB/OL].(2020-03-31) [2022-01-07] http://www.cnr.cn/yn/ynkx/20200331/t20200331_525037181.shtml
[19]	郭福涛, 苏漳文, 王光玉, 等. 我国北方针叶林人为火发生的预测模型[J]. 应用生态学报, 2015, 26(7):2099-2106.
[20]	郑凌蔚, 刘士荣, 毛军科, 等. 一种综合ANFIS和PCA的光伏发电功率预测新方法[J]. 太阳能学报, 2012, 33(6):993-1001.
[21]	李娜, 周维博, 王金凤, 等. 基于主成分分析法的ANFIS模型及其应用[J]. 人民黄河, 2014, 36(6):80-83.
[22]	王学民. 应用多元分析[M]. 上海: 上海财经大学出版社, 2010:251-280.
[23]	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.
[24]	Zhang Guoli, Wang Ming, Liu Kai. Deep neural networks for global wildfire susceptibility modelling[J]. Ecological Indicators, 2021, 127(2):107735. doi: 10.1016/j.ecolind.2021.107735
[25]	孙瑜, 史明昌, 彭欢, 等. 基于MAXENT模型的黑龙江大兴安岭森林雷击火火险预测[J]. 应用生态学报, 2014, 25(4):1100-1106.

主成分	特征值	方差贡献率/%	累计方差贡献率/%
1	3.978	44.199	44.199
2	1.922	21.354	65.554
3	1.090	12.109	77.663

变量指标	荷载值			特征向量
变量指标	1	2	3	1	2	3
x₁	-0.470	0.439	0.400	-0.236	0.317	0.383
x₂	0.352	-0.422	0.552	0.176	-0.304	0.529
x₃	0.828	0.542	0.096	0.415	0.391	0.092
x₄	-0.803	0.403	0.028	-0.403	0.291	0.027
x₅	0.809	-0.341	-0.158	0.406	-0.246	-0.151
x₆	0.612	0.700	0.256	0.307	0.505	0.245
x₇	0.868	0.434	-0.038	0.435	0.313	-0.036
x₈	0.147	-0.435	0.687	0.074	-0.314	0.658
x₉	-0.706	0.334	0.226	-0.354	0.241	0.216

因变量	变量指标	共线性统计
因变量	变量指标	容忍度	方差膨胀因子(VIF)
火灾是否发生	x₁	0.768	1.303
	x₂	0.570	1.755
	x₃	0.020	50.719
	x₄	0.287	3.487
	x₅	0.364	2.744
	x₆	0.063	15.825
	x₇	0.047	21.140
	x₈	0.828	1.208
	x₉	0.445	2.248

变量指标	样本1	样本2	样本3	样本4	样本5	全样本	样本显著量
x₁	1	1	1	0	1	1	4
x₂	1	1	1	1	1	1	5
x₄	1	1	1	1	1	1	5
x₅	0	0	1	0	0	0	1
x₈	0	0	0	0	0	0	0
x₉	1	1	1	1	1	1	5

变量指标	估计系数	标准误差	显著性(Sig)	Exp(B)
x₁	-1.557	0.693	0.025	0.211
x₂	-0.377	0.165	0.023	0.686
x₄	-0.072	0.021	0.001	0.930
x₉	-0.040	0.016	0.013	0.961
常量	6.727	1.383	0.000	835.057