|本期目录/Table of Contents|

[1]汪瑾,郭在军,业巧林,等.基于缓冲区重采样的LSTM林火预测模型*[J].中国安全生产科学技术,2023,19(2):195-202.[doi:10.11731/j.issn.1673-193x.2023.02.027]
　WANG Jin,GUO Zaijun,YE Qiaolin,et al.LSTM prediction model of forest fire based on buffer resampling[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(2):195-202.[doi:10.11731/j.issn.1673-193x.2023.02.027]
点击复制

基于缓冲区重采样的LSTM林火预测模型*

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
19
期数:
2023年2期
页码:
195-202
栏目:
职业安全卫生管理与技术
出版日期:
2023-02-28

文章信息/Info

Title:
LSTM prediction model of forest fire based on buffer resampling
文章编号:
1673-193X(2023)-02-0195-08
作者:
汪瑾; 郭在军; 业巧林; 高德民
（1.南京林业大学 信息科学技术学院，江苏 南京 210037；
2.张家口市林业科学研究院，河北 张家口 075000）
Author(s):
WANG Jin;  GUO Zaijun;  YE Qiaolin;  GAO Demin
(1.School of Information Science and Technology,Nanjing Forestry University,Nanjing Jiangsu 210037,China;
2.Zhangjiakou Forestry Research Institute,Zhangjiakou Hebei 075000,China)
关键词:
森林火灾; 风险预测; 长短期记忆模型; 时间序列
Keywords:
forest fire;  risk prediction;  long-short term memory (LSTM) model;  time series
分类号:
X954；S762
DOI:
10.11731/j.issn.1673-193x.2023.02.027
文献标志码:
A
摘要:
为提高林火风险预测精度，挖掘地图上隐含的空间信息、时间序列上隐含的长期趋势和循环波动，提出1种基于缓冲区重采样的长短期记忆（LSTM）林火预测模型，选取15个与林火相关的影响因素，以方差膨胀因子为评价指标对其进行多重共线性检验，方差膨胀因子大于10的因素具有共线性，并采用信息增益率验证筛选结果的合理性。考虑到火灾的空间聚集特性，采用缓冲区分析与过采样相结合方法减少样本不均衡现象的影响，最终得到176 732条样本。对12个影响因素和研究时间段的火点建立LSTM预测模型，对森林火灾发生风险进行预测。研究结果表明：基于缓冲区重采样的LSTM林火预测模型有效考虑时空上隐含的信息，预测模型准确率为87.06%，特异性为97.99%，敏感度为76.12%，阳性预测率为97.43%，阴性预测率为80.41%，ROC曲线与AUC值均优于随机森林（RF）和支持向量机（SVM）这2种基准算法。维尔克松秩和检验发现，本文提出的模型与基准算法结果具有显著性差异。研究结果可为提高林火风险预测精度提供参考。
Abstract:
In order to improve the prediction accuracy of forest fire risk,the spatial information hidden in the map and the long-term trend and cyclic fluctuation hidden in the time series were mined.A long-short term memory (LSTM) prediction model of forest fire based on the buffer resampling was proposed,then 15 influencing factors related to forest fire were selected,and the variance expansion factor was used as the evaluation index to carry out the multicollinearity test.The factors with variance expansion factor greater than 10 had the collinearity,and the information gain rate was used to verify the rationality of the screening results.Considering the spatial aggregation characteristics of fire,the combination method of buffer analysis and oversampling was used to reduce the impact of sample imbalance,and 176 732 samples were finally obtained.The LSTM prediction model for 12 influencing factors and fire points in the research period was established to predict the risk of forest fire.The results showed that the LSTM prediction model of forest fire based on buffer resampling effectively considered the hidden information in time and space.The accuracy of the prediction model was 87.06%,the specificity was 97.99%,the sensitivity was 76.12%,the positive prediction rate was 97.43%,and the negative prediction rate was 80.41%.Both the ROC curve and AUC value were better than the two benchmark algorithms of random forest (RF) and support vector machine (SVM).The Wilcoxon rank sum test indicated that the results of the model proposed in this study were significantly different from those of the benchmark algorithms.The research results can provide reference for improving the prediction accuracy of forest fire risk.

参考文献/References:

［1］HAO X,ZHANG G,MA S.Deep learning［J］.International Journal of Semantic Computing，2016，10(3)：417-439.
［2］HONG H Y,TSANGARATOS P,ILIA I,et al.Applying genetic algorithms to set the optimal combination of forest fire related variables and model forest fire susceptibility based on data mining models.The case of Dayu County,China［J］.Science of the Total Environment,2018，630:1044-1056.
［3］POURTAGHI Z S,POURGHASEMI H R,ARETANO R,et al.Investigation of general indicators influencing on forest fire and its susceptibility modeling using different data mining techniques［J］.Ecological Indicators,2016，64:72-84.
［4］LEUENBERGER M,PARENTE J,TONINI M,et al.Wildfire susceptibility mapping:deterministic vs.stochastic approaches［J］.Environmental Modelling & Software,2018，101:194-203.
［5］POURTAGHI Z S,POURGHASEMI H R,ARETANO R,et al.Investigation of general indicators influencing on forest fire and its susceptibility modeling using different data mining techniques［J］.Ecological Indicators,2016，64:72-84.
［6］JAFARZADEH A A,MAHDAVI A,JAFARZADEH H.Evaluation of forest fire risk using the Apriori algorithm and fuzzy c-means clustering［J］.Journal of Forest Science,2017，64:370-380.
［7］YE J,WU M,DENG Z,et al.Modeling the spatial patterns of human wildfire ignition in Yunnan province,China［J］.Applied Geography,2017，89:150-162.
［8］BUI D T,HOANG N D,SAMUI P.Spatial pattern analysis and prediction of forest fire using new machine learning approach of multivariate adaptive regression splines and differential flower pollination optimization:a case study at Laocai province (Viet Nam)［J］.Journal of Environmental Management,2019，237:476-487.
［9］DUTTA R,DAS A,ARYAL J.Big data integration shows Australian bush-fire frequency is increasing significantly［J］.Royal Society Open Science,2016，3(2):150241.
［10］HU X Y.Fire frequency analysis and prediction based on back propagation neural network model and long-short term memory mode［J］.IOP Conference Series:Earth and Environmental Science,2021，675(1):012169.
［11］MOHAN K V M,SATISH A R,RAO K M,et al.Leveraging machine learning to predict wild fires［C］//2nd International Conference on Smart Electronics and Communication,India Trichy，2021，1393-1400.
［12］侯锡铭,王伟杰,温广玉.用Logistic回归方法建立林火发生预报模型的研究［J］.东北林业大学学报,1995，23(1):102-107. HOU Ximing,WANG Weijie,WEN Guangyu.Study on establishing forecast model of forest fire occurrence using logistic regression［J］.Journal of Northeast Forestry University,1995,23(1):102-107.
［13］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.
［14］CAO Y X,WANG M,LIU K.Wildfire susceptibility assessment in Southern China:a comparison of multiple methods［J］.International Journal of Disaster Risk Science,2017,8(2):164-181.
［15］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.
［16］李永和,靳全锋.浙江省林火时空变化及影响因子分析［J］.浙江林业科技,2021，41(4):62-69. LI Yonghe,JIN Quanfeng.Temporal and spatial variation of forest fires and factors analysis in Zhejiang Province［J］.Journal of Zhejiang Forestry Science and Technology,2021，41(4):62-69.
［17］ZHANG G L,WANG M,LIU K.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.
［18］RENARD D,IRIARTE J,BIRK J J,et al.Ecological engineers ahead of their time:the functioning of pre-Columbian raised-field agriculture and its potential contributions to sustainability today［J］.Ecological Engineering,2012，45:30-44.
［19］COLKESEN I,SAHIN E K,KAVZOGLU T.Susceptibility mapping of shallow landslides using kernel-based Gaussian process,support vector machines and logistic regression［J］.Journal of African Earth Sciences,2016，118:53-64.
［20］BUDA M,MAKI A,MAZUROWSKI M A.A systematic study of the class imbalance problem in convolutional neural networks［J］.Neural Networks,2018，106:249-259.
［21］TOBLER W R.A computer movie simulating urban growth in the Detroit Region［J］.Economic Geography,1970，46(S1):234-240.

相似文献/References:

[1]王春芳,郭风平.当代森林火灾防控对策研究[J].中国安全生产科学技术,2011,7(7):168.
　Wang Chunfang,Guo Fengping.A Study on the Prevention and Control Countermeasures of Contemporary forest fires[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2011,7(2):168.
[2]程万洲,张海珊,吴超鹏,等.树冠火对输油站热辐射影响的数值模拟研究[J].中国安全生产科学技术,2012,8(10):54.
　CHENG Wan zhou,ZHANG Hai shan,WU Chao peng,et al.Modeling and numerical study on thermal radiation to oil transport station by crown fire[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2012,8(2):54.
[3]王健,武勇,刘小燕,等.景区典型树种的热解特性及动力学研究[J].中国安全生产科学技术,2019,15(2):32.[doi:10.11731/j.issn.1673-193x.2019.02.005]
　WANG Jian,WU Yong,LIU Xiaoyan,et al.Study on pyrolysis characteristics and kinetics of typical tree species in scenic spots[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2019,15(2):32.[doi:10.11731/j.issn.1673-193x.2019.02.005]
[4]汪磊,杨星月.基于蒙特卡洛方法的着陆擦机尾风险预测[J].中国安全生产科学技术,2019,15(2):45.[doi:10.11731/j.issn.1673-193x.2019.02.007]
　WANG Lei,YANG Xingyue.Risk prediction of tail strike during landing based on Monte Carlo method[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2019,15(2):45.[doi:10.11731/j.issn.1673-193x.2019.02.007]
[5]卜康正,郑先昌,沈翔,等.基坑放坡开挖对下方既有地铁隧道影响计算及风险预测[J].中国安全生产科学技术,2019,15(7):5.[doi:10.11731/j.issn.1673-193x.2019.07.001]
　BU Kangzheng,ZHENG Xianchang,SHEN Xiang,et al.Calculation for effect of foundation pit sloping excavation on underlying metro tunnels and risk prediction[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2019,15(2):5.[doi:10.11731/j.issn.1673-193x.2019.07.001]
[6]陈立林,刘琼,杜沄燕.森林自燃火灾中的凋落物辐射升温特性研究——以岳麓山为例[J].中国安全生产科学技术,2018,14(2):174.[doi:10.11731/j.issn.1673-193x.2018.02.028]
　CHEN Lilin,LIU Qiong,DU Yunyan.Research on radiant temperature rising characteristics of litter in spontaneous fire of forest——a case study of Yuelu Mountain[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(2):174.[doi:10.11731/j.issn.1673-193x.2018.02.028]
[7]马斌,张晨晨,赵怡晴,等.基于EEMD-LSTM模型的尾矿库风险预测模型研究*[J].中国安全生产科学技术,2022,18(12):116.[doi:10.11731/j.issn.1673-193x.2022.12.016]
　MA Bin,ZHANG Chenchen,ZHAO Yiqing,et al.Research on risk prediction model of tailings pond based on EEMD-LSTM model[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(2):116.[doi:10.11731/j.issn.1673-193x.2022.12.016]
[8]谢汶含,蒋永清,孙大伟,等.基于数字孪生及GA-BP神经网络的开关柜温升风险预测*[J].中国安全生产科学技术,2025,21(2):184.[doi:10.11731/j.issn.1673-193x.2025.02.024]
　XIE Wenhan,JIANG Yongqing,SUN Dawei,et al.Risk prediction on temperature rise of switchgear based on digital twin and GA-BP neural network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(2):184.[doi:10.11731/j.issn.1673-193x.2025.02.024]
[9]邓军,刘阳,康付如,等.强黏附性直链玉米淀粉基凝胶合成与阻燃性能*[J].中国安全生产科学技术,2025,21(11):75.[doi:10.11731/j.issn.1673-193x.2025.11.009]
　DENG Jun,LIU Yang,KANG Furu,et al.Synthesis and flame-retardant properties of strongly adhesive amylose-based corn-starch gel[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(2):75.[doi:10.11731/j.issn.1673-193x.2025.11.009]
[10]刘季富,刘佳奇,耿芳闯,等.中风化泥岩地层盾构结泥饼风险预测研究*[J].中国安全生产科学技术,2026,22(1):79.[doi:10.11731/j.issn.1673-193x.2026.01.010]
　LIU Jifu,LIU Jiaqi,GENG Fangchuang,et al.Clogging potential prediction for shields in moderately weathered mudstone strata[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2026,22(2):79.[doi:10.11731/j.issn.1673-193x.2026.01.010]

备注/Memo

备注/Memo:
收稿日期： 2022-05-25；网络首发日期： 2023-02-09
* 基金项目： 科技冬奥专项崇礼区森林智慧防火项目（2020SLZHFH-2）；国家自然科学基金项目（62072246）
作者简介： 汪瑾，硕士研究生，主要研究方向为时间序列林火预测。
通信作者： 高德民，博士，副教授，主要研究方向为森林防火和林业人工智能技术。
更新日期/Last Update: 2023-03-07