Article(id=1320165911189893688, tenantId=1146029695717560320, journalId=1306274063662022745, issueId=1320165840046117035, articleNumber=null, orderNo=null, doi=10.20168/j.1009-0029.2026.05.0132.11, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1739635200000, receivedDateStr=2025-02-16, revisedDate=1746115200000, revisedDateStr=2025-05-02, acceptedDate=null, acceptedDateStr=null, onlineDate=1792681680723, onlineDateStr=2026-10-22, pubDate=1778774400000, pubDateStr=2026-05-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1792681680723, onlineIssueDateStr=2026-10-22, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1792681680723, creator=13701087609, updateTime=1792681680723, updator=13701087609, issue=Issue{id=1320165840046117035, tenantId=1146029695717560320, journalId=1306274063662022745, year='2026', volume='45', issue='5', pageStart='1', pageEnd='148', issueExtLink='null', onlineDate='null', pubDate='1778774400000', pubDateStr='2026-05-15', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1792681663762, creator='13701087609', updateTime=1792681663762, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext=null, issueFiles=null, downloadFileDto=null}, startPage=132, endPage=142, ext={EN=ArticleExt(id=1320165913047970361, articleId=1320165911189893688, tenantId=1146029695717560320, journalId=1306274063662022745, language=EN, title=Research on the dual estimation method of fire state-fuel parameters for dynamic prediction of wildfires, columnId=1307033128569168044, journalTitle=Fire Science and Technology, columnName=Forest fire control, runingTitle=null, highlight=null, articleAbstract=

Since the fireline state and input parameters of wildfire spread prediction models inevitably contain errors and uncertainties, there are usually large deviations between wildfire spread predictions and actual spread results. The Kalman filter method based on data assimilation has the advantage of dynamic estimation in reducing model state errors. In this paper, we propose a dual estimation method for fireline position and fuel parameters based on Multi-Layer Perceptron (MLP) neural network, combined with the Ensemble Transform Kalman Filter (ETKF) algorithm. This method utilizes fireline position data from two consecutive observations and employs ETKF and MLP algorithms to estimate fireline position state and fuel parameters respectively under different prediction periods and fuel adjustment factor resolutions, aiming to reduce the uncertainties in both model state and model parameters. We validated the dual estimation method using a 10 000 m2 prescribed burning experiment and historical wildfire cases. The results demonstrate that the dual estimation algorithm can effectively reduce errors in both fireline state and model parameters, with the predicted fireline positions being to colser actual observations compared to single estimation methods. The prediction period significantly influences the performance of the dual estimation method in reducing fire spread model errors. The longer the prediction period, the greater the improvement in fireline position prediction accuracy compared to single estimation methods. In historical wildfire cases, the dual estimation algorithm shows good overall performance in fireline position prediction, validating the predictive capability of our algorithm in practical scenarios.

, authors=Xiaolong Luo2, Zheng Wang1, Hao Wu2, Jie Ji1, Long Ding1, authorsList=Xiaolong Luo, Zheng Wang, Hao Wu, Jie Ji, Long Ding, authorCompany=null, correspAuthors=Jie Ji, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1320165919125516887, articleId=1320165911189893688, tenantId=1146029695717560320, journalId=1306274063662022745, language=CN, title=用于森林火灾动态预测的火场状态-燃料参数双估计方法研究, columnId=1307033128950849710, journalTitle=消防科学与技术, columnName=森林火灾防控, runingTitle=null, highlight=null, articleAbstract=

由于林火蔓延预测模型的火线状态和输入参数不可避免地存在误差和不确定性,导致林火蔓延预测结果与实际蔓延结果之间通常会出现较大偏差。基于数据同化的卡尔曼滤波方法在降低模型状态误差方面具有动态估计的优势,本文结合集合转换卡尔曼滤波(Ensemble Transform Kalman Filter,ETKF)算法,提出了基于多层感知器(Multi-Layer Perceptron,MLP)神经网络的火线位置-燃料参数双估计方法。该方法利用连续两次观测到的火线位置数据,通过ETKF和MLP算法分别在不同预测周期和燃料调节因子分辨率下对火线位置状态和燃料参数进行估计,旨在降低模型状态和模型参数的不确定性。本文结合万平方米尺度的计划烧除试验和历史火灾案例,对双估计方法进行了验证。结果表明,双估计算法能够有效降低火线状态和模型参数误差,采用双估计方法预测的火线位置更接近实际观测的火线位置。预测周期对双估计方法降低火蔓延模型误差具有重要影响,预测周期越长,双估计方法相比单估计方法在火线位置预测精度上的提升越大。在历史火灾案例中,双估计算法的火线位置预测结果整体表现良好,验证了本文算法在实际场景下的预测性能。

, authors=罗肖龙2, 王正1, 吴昊2, 纪杰1, 丁龙1, authorsList=罗肖龙, 王正, 吴昊, 纪杰, 丁龙, authorCompany=null, correspAuthors=纪杰, authorNote=

罗肖龙,中国科学技术大学先进技术研究院硕士研究生,主要从事森林火灾蔓延预测研究,安徽省合肥市望江西路5089号,230026。

, correspAuthorsNote=
纪杰,中国科学技术大学研究员,博士生导师,。
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=oABbLpLomQ/c7iTevUVOUQ==, magXml=bWBidnLxPokk07frIaTVZw==, pdfUrl=null, pdf=t9CroUOiMIzfcc5Wt8Swng==, pdfFileSize=7227590, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Mqe+FvX5mhUg33ORnapiog==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=uSZfTRknhaUa2w5u9Mhn8A==, mapNumber=null, fund=null)}, authors=[Author(id=1320165919746273887, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165919821771362, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165919746273887, language=EN, stringName=Xiaolong Luo, firstName=Xiaolong, middleName=null, lastName=Luo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165919943406178, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165919746273887, language=CN, stringName=罗肖龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2中国科学技术大学 先进技术研究院,安徽 合肥 230031, bio={"content":"

罗肖龙,中国科学技术大学先进技术研究院硕士研究生,主要从事森林火灾蔓延预测研究,安徽省合肥市望江西路5089号,230026。

"}, bioImg=null, bioContent=

罗肖龙,中国科学技术大学先进技术研究院硕士研究生,主要从事森林火灾蔓延预测研究,安徽省合肥市望江西路5089号,230026。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919486227035, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=2, ext=[AuthorCompanyExt(id=1320165919494615644, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China), AuthorCompanyExt(id=1320165919503004253, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国科学技术大学 先进技术研究院,安徽 合肥 230031)])]), Author(id=1320165921755345509, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165921893757543, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165921755345509, language=EN, stringName=Zheng Wang, firstName=Zheng, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165922170581608, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165921755345509, language=CN, stringName=王正, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)])]), Author(id=1320165922241884778, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165922304799340, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922241884778, language=EN, stringName=Hao Wu, firstName=Hao, middleName=null, lastName=Wu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165922367713901, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922241884778, language=CN, stringName=吴昊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2中国科学技术大学 先进技术研究院,安徽 合肥 230031, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919486227035, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=2, ext=[AuthorCompanyExt(id=1320165919494615644, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China), AuthorCompanyExt(id=1320165919503004253, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国科学技术大学 先进技术研究院,安徽 合肥 230031)])]), Author(id=1320165922590012015, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jijie232@ustc.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1320165922682286705, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922590012015, language=EN, stringName=Jie Ji, firstName=Jie, middleName=null, lastName=Ji, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165922757784178, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922590012015, language=CN, stringName=纪杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)])]), Author(id=1320165923005248116, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165923105911414, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165923005248116, language=EN, stringName=Long Ding, firstName=Long, middleName=null, lastName=Ding, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165923168825975, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165923005248116, language=CN, stringName=丁龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)])])], keywords=[Keyword(id=1320165923525341816, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, orderNo=1, keyword=wildfire spread prediction), Keyword(id=1320165923609227897, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, orderNo=2, keyword=fuel adjustment factor), Keyword(id=1320165923873469050, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, orderNo=3, keyword=parameter estimation), Keyword(id=1320165923940577915, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, orderNo=4, keyword=ensemble transform Kalman filter algorithm), Keyword(id=1320165923999298172, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, orderNo=5, keyword=FARSITE), Keyword(id=1320165924259345021, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, orderNo=1, keyword=林火蔓延预测), Keyword(id=1320165924326453886, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, orderNo=2, keyword=燃料调节因子), Keyword(id=1320165924401951359, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, orderNo=3, keyword=参数估计), Keyword(id=1320165926046118528, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, orderNo=4, keyword=集合转换卡尔曼滤波算法), Keyword(id=1320165926134198913, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, orderNo=5, keyword=FARSITE)], refs=[Reference(id=1320165928759833248, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2024, volume=8, issue=8, pageStart=1420, pageEnd=1425, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=CUNNINGHAM C X, WILLIAMSON G J, BOWMAN D M J S, journalName=Nature Ecology & Evolution, refType=null, unstructuredReference=CUNNINGHAM C X,WILLIAMSON G J,BOWMAN D M J S.Increasing frequency and intensity of the most extreme wildfires on Earth[J].Nature Ecology & Evolution,2024,8(8):1420-1425., articleTitle=Increasing frequency and intensity of the most extreme wildfires on Earth, refAbstract=null), Reference(id=1320165928826942113, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=1998, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=1, authorNames=LINN R R, Harlow F H, journalName=null, refType=null, unstructuredReference=LINN R R,Harlow F H.Use of transport models for wildfire behavior simulations[R].Las Cruces:Los Alamos National Lab(LANL),1998., articleTitle=Use of transport models for wildfire behavior simulations, refAbstract=null), Reference(id=1320165930496275107, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2009, volume=156, issue=10, pageStart=2023, pageEnd=2041, url=null, language=null, rfNumber=3, rfOrder=2, authorNames=MELL W, MARANGHIDES A, MCDERMOTT R, journalName=Combustion and Flame, refType=null, unstructuredReference=MELL W,MARANGHIDES A,MCDERMOTT R,et al.Numerical simulation and experiments of burning Douglas fir trees[J].Combustion and Flame,2009,156(10):2023-2041., articleTitle=Numerical simulation and experiments of burning Douglas fir trees, refAbstract=null), Reference(id=1320165930588549796, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2007, volume=16, issue=1, pageStart=1, pageEnd=22, url=null, language=null, rfNumber=4, rfOrder=3, authorNames=MELL W, JENKINS M A, GOULD J, journalName=International Journal of Wildland Fire, refType=null, unstructuredReference=MELL W,JENKINS M A,GOULD J,et al.A physics-based approach to modelling grassland fires[J].International Journal of Wildland Fire,2007,16(1):1-22., articleTitle=A physics-based approach to modelling grassland fires, refAbstract=null), Reference(id=1320165930668241573, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=1983, volume=1, issue=2, pageStart=42, pageEnd=51, url=null, language=null, rfNumber=5, rfOrder=4, authorNames=王正非, journalName=山地研究, refType=null, unstructuredReference=王正非.山火初始蔓延速度测算法[J].山地研究,1983,1(2):42-51., articleTitle=山火初始蔓延速度测算法, refAbstract=null), Reference(id=1320165930743739046, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2020, volume=50, issue=1, pageStart=1, pageEnd=6,40, url=null, language=null, rfNumber=6, rfOrder=5, authorNames=张晓婷, 刘培顺, 王学芳, journalName=山东林业科技, refType=null, unstructuredReference=张晓婷,刘培顺,王学芳.王正非林火蔓延模型改进研究[J].山东林业科技,2020,50(1):1-6,40., articleTitle=王正非林火蔓延模型改进研究, refAbstract=null), Reference(id=1320165930840208039, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2022, volume=5, issue=4, pageStart=55, pageEnd=60, url=null, language=null, rfNumber=7, rfOrder=6, authorNames=王晓红, 王千雪, 于宏影, journalName=温带林业研究, refType=null, unstructuredReference=王晓红,王千雪,于宏影,等.FARSITE模型原理、结构及其在林火行为模拟中的应用[J].温带林业研究,2022,5(4):55-60., articleTitle=FARSITE模型原理、结构及其在林火行为模拟中的应用, refAbstract=null), Reference(id=1320165930970231464, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2004, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=7, authorNames=FINNEY M A, journalName=null, refType=null, unstructuredReference=FINNEY M A.FARSITE:Fire area simulator-model develop⁃ ment and evaluation[R].Ogden:USDA Forest Service Rocky⁃ Mountain Research Station,2004., articleTitle=FARSITE:Fire area simulator-model develop⁃ ment and evaluation, refAbstract=null), Reference(id=1320165931058311849, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2009, volume=18, issue=4, pageStart=349, pageEnd=368, url=null, language=null, rfNumber=9, rfOrder=8, authorNames=SULLIVAN A L, journalName=International Journal of Wildland Fire, refType=null, unstructuredReference=SULLIVAN A L.Wildland surface fire spread modelling,1990-2007.1:Physical and quasi-physical models[J].International Journal of Wildland Fire,2009,18(4):349-368., articleTitle=Wildland surface fire spread modelling,1990-2007.1:Physical and quasi-physical models, refAbstract=null), Reference(id=1320165931150586538, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2020, volume=29, issue=1, pageStart=32, pageEnd=41, url=null, language=null, rfNumber=10, rfOrder=9, authorNames=钱兰, 张启兴, 张永明, journalName=火灾科学, refType=null, unstructuredReference=钱兰,张启兴,张永明.集合卡曼滤波算法对FARSITE模型林火蔓延预测的修正效果研究[J].火灾科学,2020,29(1):32-41., articleTitle=集合卡曼滤波算法对FARSITE模型林火蔓延预测的修正效果研究, refAbstract=null), Reference(id=1320165931226084011, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=713, pageEnd=724, url=null, language=null, rfNumber=11, rfOrder=10, authorNames=ZHOU T J, JI J, JIANG Y, journalName=null, refType=null, unstructuredReference=ZHOU T J,JI J,JIANG Y,et al.EnKF-based real-time prediction of wildfire propagation[C]//The Proceedings of 11th Asia-Oceania Symposium on Fire Science and Technology.Singapore:Springer Singapore,2020:713-724., articleTitle=EnKF-based real-time prediction of wildfire propagation, refAbstract=null), Reference(id=1320165931288998572, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2019, volume=105, issue=null, pageStart=95, pageEnd=106, url=null, language=null, rfNumber=12, rfOrder=11, authorNames=ZHOU T J, DING L, JI J, journalName=Fire Safety Journal, refType=null, unstructuredReference=ZHOU T J,DING L,JI J,et al.Ensemble transform Kalman filter (ETKF) for large-scale wildland fire spread simulation using FARSITE tool and state estimation method[J].Fire Safety Journal,2019,105:95-106., articleTitle=Ensemble transform Kalman filter (ETKF) for large-scale wildland fire spread simulation using FARSITE tool and state estimation method, refAbstract=null), Reference(id=1320165931398050477, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=12, authorNames=唐纯, journalName=null, refType=null, unstructuredReference=唐纯.无人机实时动态数据驱动的林火蔓延模拟优化方法研究[D].哈尔滨:东北林业大学,2023., articleTitle=无人机实时动态数据驱动的林火蔓延模拟优化方法研究, refAbstract=null), Reference(id=1320165931460965038, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=13, authorNames=王丹, journalName=null, refType=null, unstructuredReference=王丹.林火蔓延中的数据同化方法研究[D].长沙:中南林业科技大学,2017., articleTitle=林火蔓延中的数据同化方法研究, refAbstract=null), Reference(id=1320165931519685295, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=14, authorNames=胡全, journalName=null, refType=null, unstructuredReference=胡全.基于GIS的森林火场模拟关键技术研究[D].哈尔滨:东北林业大学,2014., articleTitle=基于GIS的森林火场模拟关键技术研究, refAbstract=null), Reference(id=1320165931586794160, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=15, authorNames=汪瑾, journalName=null, refType=null, unstructuredReference=汪瑾.基于长短期记忆网络的林火风险与蔓延预测研究[D].南京:南京林业大学,2023., articleTitle=基于长短期记忆网络的林火风险与蔓延预测研究, refAbstract=null), Reference(id=1320165931653903025, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2020, volume=116, issue=null, pageStart=103167, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=16, authorNames=ZHOU T J, DING L, JI J, journalName=Fire Safety Journal, refType=null, unstructuredReference=ZHOU T J,DING L,JI J,et al.Combined estimation of fire perimeters and fuel adjustment factors in FARSITE for forecasting wildland fire propagation[J].Fire Safety Journal,2020,116:103167., articleTitle=Combined estimation of fire perimeters and fuel adjustment factors in FARSITE for forecasting wildland fire propagation, refAbstract=null), Reference(id=1320165931725206194, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2024, volume=6, issue=2, pageStart=125, pageEnd=138, url=null, language=null, rfNumber=18, rfOrder=17, authorNames=PRASAD D V, SARASWATHI S, journalName=Journal of Trends in Computer Science and Smart Technology, refType=null, unstructuredReference=PRASAD D V ,SARASWATHI S.The role of anomaly detection in industry 4.0:A survey of techniques and applications[J].Journal of Trends in Computer Science and Smart Technology,2024,6(2):125-138., articleTitle=The role of anomaly detection in industry 4.0:A survey of techniques and applications, refAbstract=null), Reference(id=1320165931809092275, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=11, pageStart=66, pageEnd=70, url=null, language=null, rfNumber=19, rfOrder=18, authorNames=谢洋, 陆新建, 邵晓雯, journalName=化学工程, refType=null, unstructuredReference=谢洋,陆新建,邵晓雯,等.基于MLP-RF组合模型预测加氢裂化装置柴油产量[J].化学工程,2024,52(11):66-70., articleTitle=基于MLP-RF组合模型预测加氢裂化装置柴油产量, refAbstract=null), Reference(id=1320165931876201140, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=9, url=null, language=null, rfNumber=20, rfOrder=19, authorNames=何卫, 牛岱松, 宋世琦, journalName=振动工程学报, refType=null, unstructuredReference=何卫,牛岱松,宋世琦.实测数据驱动的地铁振动源强MLP神经网络预测方法[J/OL].振动工程学报,1-9., articleTitle=实测数据驱动的地铁振动源强MLP神经网络预测方法, refAbstract=null), Reference(id=1320165931939115701, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2025, volume=37, issue=10, pageStart=2568, pageEnd=2577, url=null, language=null, rfNumber=21, rfOrder=20, authorNames=黄德启, 涂亚婷, 张振华, journalName=系统仿真学报, refType=null, unstructuredReference=黄德启,涂亚婷,张振华,等.基于MLP与改进GCN-TD3的交通信号控制建模与仿真[J].系统仿真学报,2025,37(10):2568-2577., articleTitle=基于MLP与改进GCN-TD3的交通信号控制建模与仿真, refAbstract=null), Reference(id=1320165932006224566, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2023, volume=40, issue=10, pageStart=233, pageEnd=240, url=null, language=null, rfNumber=22, rfOrder=21, authorNames=武慧荣, 陈少阳, 崔淑华, journalName=公路交通科技, refType=null, unstructuredReference=武慧荣,陈少阳,崔淑华.基于GM(1,1)-MLP神经网络模型的大宗货物运输需求预测[J].公路交通科技,2023,40(10):233-240., articleTitle=基于GM(1,1)-MLP神经网络模型的大宗货物运输需求预测, refAbstract=null), Reference(id=1320165932077527735, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2024, volume=39, issue=6, pageStart=715, pageEnd=725, url=null, language=null, rfNumber=23, rfOrder=22, authorNames=茅晓晨, 焦英辰, 陈璇, journalName=海军航空大学学报, refType=null, unstructuredReference=茅晓晨,焦英辰,陈璇,等.基于POD和多层感知器的跨声速压气机叶型精细优化设计方法[J].海军航空大学学报,2024,39(6):715-725., articleTitle=基于POD和多层感知器的跨声速压气机叶型精细优化设计方法, refAbstract=null), Reference(id=1320165932228522680, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=49, pageEnd=53, url=null, language=null, rfNumber=24, rfOrder=23, authorNames=窦淼, 侯祥宁, journalName=水利信息化, refType=null, unstructuredReference=窦淼,侯祥宁.PCA-MLP神经网络模型在黄河宁夏段径流预测中的应用[J].水利信息化,2024(4):49-53., articleTitle=PCA-MLP神经网络模型在黄河宁夏段径流预测中的应用, refAbstract=null), Reference(id=1320165932304020153, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=5, pageStart=1169, pageEnd=1176, url=null, language=null, rfNumber=25, rfOrder=24, authorNames=肖舜祯, 刘强, 徐志扬, journalName=江西农业大学学报, refType=null, unstructuredReference=肖舜祯,刘强,徐志扬,等.基于MLP的上海市主要树种单木胸径生长率模型[J].江西农业大学学报,2022,44(5):1169-1176., articleTitle=基于MLP的上海市主要树种单木胸径生长率模型, refAbstract=null), Reference(id=1320165932408877754, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2021, volume=30, issue=10, pageStart=776, pageEnd=789, url=null, language=null, rfNumber=26, rfOrder=25, authorNames=TANEJA R, HILTON J, WALLACE L, journalName=International Journal of Wildland Fire, refType=null, unstructuredReference=TANEJA R,HILTON J,WALLACE L,et al.Effect of fuel spatial resolution on predictive wildfire models[J].International Journal of Wildland Fire,2021,30(10):776-789., articleTitle=Effect of fuel spatial resolution on predictive wildfire models, refAbstract=null), Reference(id=1320165932475986619, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=26, authorNames=张时雨, journalName=null, refType=null, unstructuredReference=张时雨.基于机载监测系统的东北森林地表火蔓延与预测模型研究[D].哈尔滨:东北林业大学,2022., articleTitle=基于机载监测系统的东北森林地表火蔓延与预测模型研究, refAbstract=null), Reference(id=1320165932568261308, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=27, authorNames=胡宏宇, journalName=null, refType=null, unstructuredReference=胡宏宇.基于MLP的海上无人跨域协同效能评估系统研究与实现[D].沈阳:辽宁大学,2024., articleTitle=基于MLP的海上无人跨域协同效能评估系统研究与实现, refAbstract=null), Reference(id=1320165932639564477, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=2023, volume=8, issue=2, pageStart=56, pageEnd=62, url=null, language=null, rfNumber=29, rfOrder=28, authorNames=YINKA-BANJO C O, AKINYEMI M I, YAMA B B, journalName=Annals of Science and Technology, refType=null, unstructuredReference=YINKA-BANJO C O,AKINYEMI M I,YAMA B B.Ensemble algorithm for simulated corrosion data-tentative[J].Annals of Science and Technology,2023,8(2):56-62., articleTitle=Ensemble algorithm for simulated corrosion data-tentative, refAbstract=null), Reference(id=1320165932715061950, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, doi=null, pmid=null, pmcid=null, year=1969, volume=43, issue=5, pageStart=1130, pageEnd=1136, url=null, language=null, rfNumber=30, rfOrder=29, authorNames=CHEETHAM A H, HAZEL J E, journalName=Journal of Paleontology, refType=null, unstructuredReference=CHEETHAM A H,HAZEL J E.Binary (presence-absence) similarity coefficients[J].Journal of Paleontology,1969,43(5):1130-1136., articleTitle=Binary (presence-absence) similarity coefficients, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)]), AuthorCompany(id=1320165919486227035, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=2, ext=[AuthorCompanyExt(id=1320165919494615644, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China), AuthorCompanyExt(id=1320165919503004253, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国科学技术大学 先进技术研究院,安徽 合肥 230031)])], figs=[ArticleFig(id=1320165926297776770, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.1, caption=Flow chart of the dual estimation algorithm for fireline position-fuel conditioning factor, figureFileSmall=9/4ns3zm6UNxGVQZC+g9jQ==, figureFileBig=Mqe+FvX5mhUg33ORnapiog==, tableContent=null), ArticleFig(id=1320165926406828675, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图1, caption=火线位置-燃料调节因子双估计算法流程图, figureFileSmall=9/4ns3zm6UNxGVQZC+g9jQ==, figureFileBig=Mqe+FvX5mhUg33ORnapiog==, tableContent=null), ArticleFig(id=1320165926599766660, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.2, caption=The root mean square error between the predicted results and the real values of different set size dual estimation algorithms, figureFileSmall=KJIUvzBJdM8AAI1Y4+OcxQ==, figureFileBig=5xQzcY0IhXNrxK5jl6B2kQ==, tableContent=null), ArticleFig(id=1320165926696235653, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图2, caption=不同集合大小双估计算法预测结果与真实值之间的均方根误差值, figureFileSmall=KJIUvzBJdM8AAI1Y4+OcxQ==, figureFileBig=5xQzcY0IhXNrxK5jl6B2kQ==, tableContent=null), ArticleFig(id=1320165926767538822, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.3, caption=Planned burning experimental site and experimental instrumentation setup, figureFileSmall=8SDGRZMKgZ/a9uo0B/xSyw==, figureFileBig=a41AfLgIaFK66I1dUtjJzA==, tableContent=null), ArticleFig(id=1320165926859813511, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图3, caption=计划烧除试验场地及试验仪器设置, figureFileSmall=8SDGRZMKgZ/a9uo0B/xSyw==, figureFileBig=a41AfLgIaFK66I1dUtjJzA==, tableContent=null), ArticleFig(id=1320165926926922376, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.4, caption=Planned burning experimental site and position of the fireline at different times, figureFileSmall=bep6Ok8h2jhpEboDDdCbWA==, figureFileBig=2LPiA3MdAs4gM4rO+HzdYg==, tableContent=null), ArticleFig(id=1320165927015002761, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图4, caption=计划烧除试验场地及不同时刻火线位置, figureFileSmall=bep6Ok8h2jhpEboDDdCbWA==, figureFileBig=2LPiA3MdAs4gM4rO+HzdYg==, tableContent=null), ArticleFig(id=1320165927082111626, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.5, caption=Resolution of different fuel conditioning factors in planned burning experiments, figureFileSmall=uwTMNOIUI/0WZ7LJwntwng==, figureFileBig=Dq3XtP8vPvtgrVWVust5Zg==, tableContent=null), ArticleFig(id=1320165927157609099, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图5, caption=计划烧除试验中不同燃料调节因子分辨率, figureFileSmall=uwTMNOIUI/0WZ7LJwntwng==, figureFileBig=Dq3XtP8vPvtgrVWVust5Zg==, tableContent=null), ArticleFig(id=1320165927233106572, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.6, caption=Comparison of fireline positions and similarity index SI between single and dual estimation predictions when the prediction period is 2 min, figureFileSmall=bzEOavPuRpYRWTlMCwgc5g==, figureFileBig=RMpfW9RzGU9BTQg7cCbxWw==, tableContent=null), ArticleFig(id=1320165927300215437, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图6, caption=预测周期为2 min时单、双估计预测的火线位置及相似性指数SI对比, figureFileSmall=bzEOavPuRpYRWTlMCwgc5g==, figureFileBig=RMpfW9RzGU9BTQg7cCbxWw==, tableContent=null), ArticleFig(id=1320165927367324302, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.7, caption=Average similarity index SI and prediction accuracy of single and dual estimation methods under different prediction cycles, figureFileSmall=fewUTzbsFpo7zpkbFu/aVg==, figureFileBig=PVkF7TJHSX6C7tpXh22MmQ==, tableContent=null), ArticleFig(id=1320165927434433167, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图7, caption=不同预测周期下单、双估计方法平均相似性指数SI及预测精度, figureFileSmall=fewUTzbsFpo7zpkbFu/aVg==, figureFileBig=PVkF7TJHSX6C7tpXh22MmQ==, tableContent=null), ArticleFig(id=1320165927493153424, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.8, caption=Single and dual estimated mean similarity index SI and prediction accuracy at different fuel conditioning factor resolutions, figureFileSmall=SKL21cgnGaDu4KRrDhO+aA==, figureFileBig=fWs2/yeJ2CieugsZzm5EPw==, tableContent=null), ArticleFig(id=1320165927551873681, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图8, caption=不同燃料调节因子分辨率下单、双估计平均相似性指数SI及预测精度, figureFileSmall=SKL21cgnGaDu4KRrDhO+aA==, figureFileBig=fWs2/yeJ2CieugsZzm5EPw==, tableContent=null), ArticleFig(id=1320165927627371154, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.9, caption=Comparison of predicted and observed fireline positions and fuel conditioning factor values for single and dual estimation methods at a fuel conditioning resolution of 200 m and a prediction period of 2 min, figureFileSmall=ss1Cz6EqxW9eBWrvtfK6pg==, figureFileBig=pxxulkXLetSB8l/xfZjRqQ==, tableContent=null), ArticleFig(id=1320165927690285715, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图9, caption=燃料调节分辨率为200 m、预测周期为2 min时单、双估计方法预测火线位置与观测火线位置及燃料调节因子值对比, figureFileSmall=ss1Cz6EqxW9eBWrvtfK6pg==, figureFileBig=pxxulkXLetSB8l/xfZjRqQ==, tableContent=null), ArticleFig(id=1320165927782560404, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.10, caption=Map of actual observed fireline locations for different forecast cycles, figureFileSmall=oqweokBc718GmXTCjjvXkQ==, figureFileBig=aJZy1NU1XxWUwRNfoq0jdw==, tableContent=null), ArticleFig(id=1320165927870640789, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图10, caption=不同预测周期下实际的观测火线位置图, figureFileSmall=oqweokBc718GmXTCjjvXkQ==, figureFileBig=aJZy1NU1XxWUwRNfoq0jdw==, tableContent=null), ArticleFig(id=1320165927954526870, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.11, caption=Terrain feature information of historical fire cases, figureFileSmall=w2uYUX1vpgVOpcuGBJYDcQ==, figureFileBig=nbnXZ9t91OsKo5WVJ2h1vQ==, tableContent=null), ArticleFig(id=1320165928038412951, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图11, caption=历史火灾案例地形特征信息, figureFileSmall=w2uYUX1vpgVOpcuGBJYDcQ==, figureFileBig=nbnXZ9t91OsKo5WVJ2h1vQ==, tableContent=null), ArticleFig(id=1320165928139076248, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.12, caption=Meteorological information of historical fire cases, figureFileSmall=UT39juEowvJG+r0I0y7yOw==, figureFileBig=9zO2igDBl7+ayucoqa83zg==, tableContent=null), ArticleFig(id=1320165928214573721, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图12, caption=历史火灾案例的气象信息, figureFileSmall=UT39juEowvJG+r0I0y7yOw==, figureFileBig=9zO2igDBl7+ayucoqa83zg==, tableContent=null), ArticleFig(id=1320165928294265498, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Fig.13, caption=Prediction results of fireline position in historical fire cases, figureFileSmall=REjMhXAVYNDBJkNe8CUCPg==, figureFileBig=/B1JM4QWVtvrEd04/+yulg==, tableContent=null), ArticleFig(id=1320165928369762971, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=图13, caption=历史火灾案例的火线位置预测结果, figureFileSmall=REjMhXAVYNDBJkNe8CUCPg==, figureFileBig=/B1JM4QWVtvrEd04/+yulg==, tableContent=null), ArticleFig(id=1320165928436871836, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Table 1, caption=

Input data for meteorological parameters for planned burning experiments

, figureFileSmall=null, figureFileBig=null, tableContent=
时间风速/km/h风向
(顺时针从正北开始)/(°)
11:0830283
11:0939283
11:1029286
11:1115288
11:1225285
11:1325328
11:1417310
11:1514295
11:1625335
11:1731285
11:1833284
11:1927291
11:2020302
11:2116290
11:2226289
11:2324312
11:2418295
11:2517294
), ArticleFig(id=1320165928503980701, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=表1, caption=

计划烧除试验气象参数的输入数据

, figureFileSmall=null, figureFileBig=null, tableContent=
时间风速/km/h风向
(顺时针从正北开始)/(°)
11:0830283
11:0939283
11:1029286
11:1115288
11:1225285
11:1325328
11:1417310
11:1514295
11:1625335
11:1731285
11:1833284
11:1927291
11:2020302
11:2116290
11:2226289
11:2324312
11:2418295
11:2517294
), ArticleFig(id=1320165928575283870, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=EN, label=Table 2, caption=

Design of experimental conditions for planned burning

, figureFileSmall=null, figureFileBig=null, tableContent=
燃料调节因子估计值预测周期/min燃料调节因子分辨率/m
0.6、3.0、5.0、7.0、9.02、3、4200、100、50、25
), ArticleFig(id=1320165928634004127, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, language=CN, label=表2, caption=

计划烧除试验工况设计

, figureFileSmall=null, figureFileBig=null, tableContent=
燃料调节因子估计值预测周期/min燃料调节因子分辨率/m
0.6、3.0、5.0、7.0、9.02、3、4200、100、50、25
)], attaches=null, journal=Journal(id=1306266559758565460, delFlag=0, nameCn=消防科学与技术, nameEn=Fire Science and Technology, nameHistory1=null, nameHistory2=null, issn=1009-0029, eissn=null, cn=12-1311/TU, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=QeUyB10QrUvK+bUQ0VpXvA==, journalPrice=null, startedYear=null, abbrevIsoEn=Fire Science and Technology, journalRemark=null, publicationField=null, createdTime=1789367817107, updatedTime=1789369955418, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=X, firstLetterEn=X, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=QeUyB10QrUvK+bUQ0VpXvA==, picEn=HN5KLNjdXUtcFai/4GlSSw==, jcr=null, cjcr=null, exts=[JournalExt(id=1306275528554607205, language=CN, name=消防科学与技术, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1789369955433, updatedTime=1789369955433, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://xfkxyjsauthor.manuscriptcloud.com/, submissionEditorUrl=https://xfkxyjseditor.manuscriptcloud.com/, submissionReviewUrl=https://xfkxyjsauthor.manuscriptcloud.com/, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1306275528684630630, language=EN, name=Fire Science and Technology, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1789369955464, updatedTime=1789369955464, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://xfkxyjsauthor.manuscriptcloud.com/, submissionEditorUrl=https://xfkxyjseditor.manuscriptcloud.com/, submissionReviewUrl=https://xfkxyjsauthor.manuscriptcloud.com/, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1306274063662022745, websiteList=[Website(id=1306284065322197953, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1306274063662022745, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/xfkxyjs/CN, language=CN, createTime=1789371990757, createBy=18614031015, updateTime=1789372713663, updateBy=18614031015, name=消防科学与技术-中文, tplId=1146099689490845704, title=消防科学与技术, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1306287234127258043, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=articleTextType, value=kx, createTime=1789372746259, updateTime=1789372746259, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234106286520, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=banner, value=null, createTime=1789372746254, updateTime=1789372746254, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234156618174, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=grayFlag, value=0, createTime=1789372746266, updateTime=1789372746266, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234093703607, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=logo, value=https://castjournals.cast.org.cn/joweb/xfkxyjs/CN/file/pic?fileId=eMnTNnjXjUQVmppsrxDsRQ==, createTime=1789372746251, updateTime=1789372746251, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234173395392, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=minRunFlag, value=0, createTime=1789372746270, updateTime=1789372746270, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234118869434, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/xfkxyjs/CN/file/pic, createTime=1789372746257, updateTime=1789372746257, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234165006783, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=silenceFlag, value=0, createTime=1789372746268, updateTime=1789372746268, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234110480825, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1789372746255, updateTime=1789372746255, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234135646652, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=themeColor, value=null, createTime=1789372746261, updateTime=1789372746261, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287234148229565, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065322197953, code=themeStyle, value=null, createTime=1789372746264, updateTime=1789372746264, creator=18614031015, updator=18614031015)]), Website(id=1306284065737434056, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1306274063662022745, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/xfkxyjs/EN, language=EN, createTime=1789371990856, createBy=18614031015, updateTime=1789372692991, updateBy=18614031015, name=消防科学与技术-英文, tplId=1146101810881728533, title=Fire Science and Technology, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1306287202959385009, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=articleTextType, value=kx, createTime=1789372738828, updateTime=1789372738828, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202925830574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=banner, value=null, createTime=1789372738820, updateTime=1789372738820, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202980356532, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=grayFlag, value=0, createTime=1789372738833, updateTime=1789372738833, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202913247661, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=logo, value=https://castjournals.cast.org.cn/joweb/xfkxyjs/EN/file/pic?fileId=eMnTNnjXjUQVmppsrxDsRQ==, createTime=1789372738817, updateTime=1789372738817, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202992939446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=minRunFlag, value=0, createTime=1789372738836, updateTime=1789372738836, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202950996400, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/xfkxyjs/EN/file/pic, createTime=1789372738826, updateTime=1789372738826, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202984550837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=silenceFlag, value=0, createTime=1789372738834, updateTime=1789372738834, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202938413487, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1789372738823, updateTime=1789372738823, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202967773618, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=themeColor, value=null, createTime=1789372738830, updateTime=1789372738830, creator=18614031015, updator=18614031015), WebsiteProps(id=1306287202971967923, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1306284065737434056, code=themeStyle, value=null, createTime=1789372738831, updateTime=1789372738831, creator=18614031015, updator=18614031015)])], journalTitle=消防科学与技术, weixinUrl=null, journalUrl=https://www.xfkj.com.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Fire Science and Technology, journalPhotoCn=QeUyB10QrUvK+bUQ0VpXvA==, journalPhotoEn=HN5KLNjdXUtcFai/4GlSSw==, journalFirstLetter=X, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/xfkxyjs/CN/10.20168/j.1009-0029.2026.05.0132.11, detailUrlEn=https://castjournals.cast.org.cn/joweb/xfkxyjs/EN/10.20168/j.1009-0029.2026.05.0132.11, pdfUrlCn=https://castjournals.cast.org.cn/joweb/xfkxyjs/CN/PDF/10.20168/j.1009-0029.2026.05.0132.11, pdfUrlEn=https://castjournals.cast.org.cn/joweb/xfkxyjs/EN/PDF/10.20168/j.1009-0029.2026.05.0132.11, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1778774400000, fullTextJson=null, articleText=null, reference=null)
收藏切换
用于森林火灾动态预测的火场状态-燃料参数双估计方法研究
收藏切换
PDF下载
罗肖龙 2 , 王正 1 , 吴昊 2 , 纪杰 1 , 丁龙 1
消防科学与技术 | 森林火灾防控 2026,45(5): 132-142
收起
收藏切换
消防科学与技术 |森林火灾防控 2026 , 45 (5) : 132 -142
用于森林火灾动态预测的火场状态-燃料参数双估计方法研究
全屏
[Author(id=1320165919746273887, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165919821771362, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165919746273887, language=EN, stringName=Xiaolong Luo, firstName=Xiaolong, middleName=null, lastName=Luo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165919943406178, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165919746273887, language=CN, stringName=罗肖龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2中国科学技术大学 先进技术研究院,安徽 合肥 230031, bio={"content":"

罗肖龙,中国科学技术大学先进技术研究院硕士研究生,主要从事森林火灾蔓延预测研究,安徽省合肥市望江西路5089号,230026。

"}, bioImg=null, bioContent=

罗肖龙,中国科学技术大学先进技术研究院硕士研究生,主要从事森林火灾蔓延预测研究,安徽省合肥市望江西路5089号,230026。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919486227035, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=2, ext=[AuthorCompanyExt(id=1320165919494615644, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China), AuthorCompanyExt(id=1320165919503004253, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国科学技术大学 先进技术研究院,安徽 合肥 230031)])]), Author(id=1320165921755345509, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165921893757543, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165921755345509, language=EN, stringName=Zheng Wang, firstName=Zheng, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165922170581608, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165921755345509, language=CN, stringName=王正, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)])]), Author(id=1320165922241884778, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165922304799340, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922241884778, language=EN, stringName=Hao Wu, firstName=Hao, middleName=null, lastName=Wu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165922367713901, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922241884778, language=CN, stringName=吴昊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2中国科学技术大学 先进技术研究院,安徽 合肥 230031, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919486227035, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=2, ext=[AuthorCompanyExt(id=1320165919494615644, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China), AuthorCompanyExt(id=1320165919503004253, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919486227035, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2中国科学技术大学 先进技术研究院,安徽 合肥 230031)])]), Author(id=1320165922590012015, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jijie232@ustc.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1320165922682286705, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922590012015, language=EN, stringName=Jie Ji, firstName=Jie, middleName=null, lastName=Ji, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165922757784178, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165922590012015, language=CN, stringName=纪杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)])]), Author(id=1320165923005248116, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320165923105911414, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165923005248116, language=EN, stringName=Long Ding, firstName=Long, middleName=null, lastName=Ding, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320165923168825975, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, authorId=1320165923005248116, language=CN, stringName=丁龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320165919406535256, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, xref=1, ext=[AuthorCompanyExt(id=1320165919414923865, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China), AuthorCompanyExt(id=1320165919423312474, tenantId=1146029695717560320, journalId=1306274063662022745, articleId=1320165911189893688, companyId=1320165919406535256, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026)])])]
罗肖龙2, 王正1, 吴昊2, 纪杰1 , 丁龙1
作者信息
  • 1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026
  • 2中国科学技术大学 先进技术研究院,安徽 合肥 230031
通讯作者:
纪杰,中国科学技术大学研究员,博士生导师,。
作者简介:

罗肖龙,中国科学技术大学先进技术研究院硕士研究生,主要从事森林火灾蔓延预测研究,安徽省合肥市望江西路5089号,230026。

Research on the dual estimation method of fire state-fuel parameters for dynamic prediction of wildfires
Xiaolong Luo2, Zheng Wang1, Hao Wu2, Jie Ji1 , Long Ding1
Affiliations
  • 1State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei Anhui 230026, China
  • 2Institute of Advanced Technology, University of Science and Technology of China, Hefei Anhui 230031, China
出版时间: 2026-05-15 doi: 10.20168/j.1009-0029.2026.05.0132.11
文章导航
收藏切换

由于林火蔓延预测模型的火线状态和输入参数不可避免地存在误差和不确定性,导致林火蔓延预测结果与实际蔓延结果之间通常会出现较大偏差。基于数据同化的卡尔曼滤波方法在降低模型状态误差方面具有动态估计的优势,本文结合集合转换卡尔曼滤波(Ensemble Transform Kalman Filter,ETKF)算法,提出了基于多层感知器(Multi-Layer Perceptron,MLP)神经网络的火线位置-燃料参数双估计方法。该方法利用连续两次观测到的火线位置数据,通过ETKF和MLP算法分别在不同预测周期和燃料调节因子分辨率下对火线位置状态和燃料参数进行估计,旨在降低模型状态和模型参数的不确定性。本文结合万平方米尺度的计划烧除试验和历史火灾案例,对双估计方法进行了验证。结果表明,双估计算法能够有效降低火线状态和模型参数误差,采用双估计方法预测的火线位置更接近实际观测的火线位置。预测周期对双估计方法降低火蔓延模型误差具有重要影响,预测周期越长,双估计方法相比单估计方法在火线位置预测精度上的提升越大。在历史火灾案例中,双估计算法的火线位置预测结果整体表现良好,验证了本文算法在实际场景下的预测性能。

林火蔓延预测  /  燃料调节因子  /  参数估计  /  集合转换卡尔曼滤波算法  /  FARSITE

Since the fireline state and input parameters of wildfire spread prediction models inevitably contain errors and uncertainties, there are usually large deviations between wildfire spread predictions and actual spread results. The Kalman filter method based on data assimilation has the advantage of dynamic estimation in reducing model state errors. In this paper, we propose a dual estimation method for fireline position and fuel parameters based on Multi-Layer Perceptron (MLP) neural network, combined with the Ensemble Transform Kalman Filter (ETKF) algorithm. This method utilizes fireline position data from two consecutive observations and employs ETKF and MLP algorithms to estimate fireline position state and fuel parameters respectively under different prediction periods and fuel adjustment factor resolutions, aiming to reduce the uncertainties in both model state and model parameters. We validated the dual estimation method using a 10 000 m2 prescribed burning experiment and historical wildfire cases. The results demonstrate that the dual estimation algorithm can effectively reduce errors in both fireline state and model parameters, with the predicted fireline positions being to colser actual observations compared to single estimation methods. The prediction period significantly influences the performance of the dual estimation method in reducing fire spread model errors. The longer the prediction period, the greater the improvement in fireline position prediction accuracy compared to single estimation methods. In historical wildfire cases, the dual estimation algorithm shows good overall performance in fireline position prediction, validating the predictive capability of our algorithm in practical scenarios.

wildfire spread prediction  /  fuel adjustment factor  /  parameter estimation  /  ensemble transform Kalman filter algorithm  /  FARSITE
罗肖龙, 王正, 吴昊, 纪杰, 丁龙. 用于森林火灾动态预测的火场状态-燃料参数双估计方法研究. 消防科学与技术, 2026 , 45 (5) : 132 -142 . DOI: 10.20168/j.1009-0029.2026.05.0132.11
Xiaolong Luo, Zheng Wang, Hao Wu, Jie Ji, Long Ding. Research on the dual estimation method of fire state-fuel parameters for dynamic prediction of wildfires[J]. Fire Science and Technology, 2026 , 45 (5) : 132 -142 . DOI: 10.20168/j.1009-0029.2026.05.0132.11
随着人类活动的加剧,全球气候变得更加干燥炎热,森林火灾发生的频率及强度显著增加[1]。为了服务于森林火灾应急管理决策,研究人员开发了多种火灾蔓延预测工具,主要分为以下3类:基于物理规律建模(如FIRETEC[2-3]和WFDS[4]);基于经验规律建模(如王正非林火蔓延预测模型[5-6]);基于半物理半经验规律建模(如FARSITE[7-8])。对于基于明确物理规律建立的火灾蔓延预测工具,其预测所需的计算时间较长,且可操作性往往不能满足突发森林火灾紧急情况下的快速预测需求[9]。因此,研究人员更多地使用第2和第3类林火蔓延预测模型。然而,基于经验规律或半物理半经验规律建立的模型主要使用简化的物理函数,并结合实验室火行为试验数据统计得出的校准常数进行构建。此外,上述模型的输入参数通常是静态不变的。因此,将这些预测模型应用于实际的林火场景时会产生较大的误差累积。根据误差来源可以分为状态误差和参数误差。状态误差是指受建模规律限制的林火蔓延预测模型在预测现实中某一时刻的火灾状态时(如火灾位置、火灾蔓延速率及火灾反应强度等)所产生的预测误差。参数误差是指由于在整个火灾发展过程中未能及时更新模型输入参数(如环境风速、环境湿度及可燃物含水率等)而导致的预测误差。
在降低林火蔓延预测模型状态误差方面,针对FARSITE中观测火线位置输入存在偏差的问题,钱兰等[10]以及ZHOU T J等[11]利用在预测过程中不断融入新观测数据的数据同化思想,提出了一种基于集合卡尔曼滤波算法的动态修正方法。ZHOU T J等[12]进一步提出了基于集合转换卡尔曼滤波算法(Ensemble Transform Kalman Filter,ETKF)的火线位置动态估计方法,有效降低了林火蔓延预测模型中的状态误差。在本文中,将上述基于数据同化方法仅修正每一模拟时间步长的初始火线位置的方法称为“单估计方法”。虽然上述单估计方法能够修正火灾状态,但在实际林火场景中,输入参数(如可燃物参数等)是不断变化的,而单估计方法并未对输入参数进行实时动态校正。随着预测时长的增加,参数误差的不断累积将导致预测精度下降。因此,单估计方法仍存在一定局限性。在降低林火蔓延预测模型参数误差方面,唐纯[13]以及王丹[14]通过野外点火试验获取了火蔓延速度与风矢量的对应关系,在预测过程中利用神经网络算法不断估计并调整风矢量,从而降低火蔓延预测模型的参数误差。胡全[15]以及汪瑾[16]分别基于长短期记忆神经网络算法和王正非林火蔓延模型,结合室内点火试验,研究了不同地形坡度、环境风速及可燃物含水率和载量等因素对森林地表火蔓延预测的影响,并据此提出了CA-SDP模型和修正的王正非林火蔓延模型。
上述较多文献表明,数据同化ETKF算法在降低林火蔓延预测模型的状态误差上具有较好的效果。在降低模型参数误差方面,神经网络算法(如RBFNN、LSTM、Transformer等算法)常被用于对模型输入参数进行回归预测。例如,ZHOU T J等[17]使用径向基神经网络算法对燃料调节因子进行估计,从而降低了模型参数误差。但径向基神经网络算法对噪声数据较为敏感且其性能高度依赖于径向基函数的选择。LSTM通过门控机制(遗忘门、输入门、输出门)解决了传统RNN的梯度消失问题,擅长捕捉时间序列中的长期依赖关系,在时序预测(如能源产量预测、气象数据建模)中表现突出。然而,LSTM需要大量高质量数据进行训练,小数据集易导致过拟合[18]。Transformer基于自注意力机制,通过并行计算全局依赖关系,在长序列任务(如机器翻译、文本生成等)中表现优异,但其自注意力的时间复杂度为 O(n2),当输入维度较高时,计算资源消耗显著增加。尽管LSTM与Transformer在时序建模中表现卓越,但其结构复杂性及计算成本在参数估计任务中可能成为瓶颈。MLP(Multi-Layer Perceptron,MLP)凭借其简洁的全连接架构、高效的非线性拟合能力及抗噪性,更适合燃料调节因子的回归预测需求。同时,MLP神经网络具有较强的灵活性和可扩展性,可通过增加隐藏层和神经元的数量提高模型的复杂度和表达能力,这使其在分类、回归以及更广泛的机器学习任务中得到广泛应用,例如在能源产量预测[19]、交通运输[20-22]、工业设计[23]以及环境保护[24-25]等领域。在双估计方法中,ETKF与MLP的结合可以更快速地估计出某一时刻下火线的分析值(状态估计)及最大相似性指数对应的燃料调节因子值(参数估计),二者结合可减小林火蔓延预测过程中的状态误差和参数误差,实现更快速、更准确的火线位置预测。因此,本文结合ETKF单估计方法,提出基于MLP神经网络算法的火线位置-燃料参数双估计方法,研究不同预测周期和燃料调节因子分辨率下双估计方法在降低林火蔓延预测模型的状态误差和参数误差方面的效果。
此外,上述研究尚未深入讨论预测模型的时间分辨率和参数空间分辨率对预测结果的影响。最近,TANEJA R等[26]利用机载激光扫描数据,系统评估了不同空间分辨率(2、5、7、10、30、50 m)的燃料数据对Mallee林火蔓延预测模型行为的影响,但其研究主要基于Mallee植被类型,并未测试不同环境下燃料空间分辨率与火灾行为之间的关系。张时雨[27]结合简化的Rothermel模型、元胞自动机以及数据同化算法,提出了多向性MCA-ENKF森林地表火蔓延预测模型,研究并确定了最佳的集合个数和预测周期组合,通过室内点火试验验证了其模型在不同环境下的预测效果。该研究确定的最佳集合个数和预测周期组合主要是为其提出的新林火蔓延预测模型提供较好的初始化参数,并未详细讨论不同集合个数和预测周期组合对降低模型误差的影响。文献[17]基于数据同化算法和径向基神经网络算法,提出了一种同时估计状态误差和参数误差的双估计方法。该研究指出,延长预测周期会导致算法预测精度下降,但并未继续深入分析最优预测周期的选择问题。预测周期和燃料调节因子分辨率作为林火蔓延预测模型中的重要初始化参数,其不同的选择往往决定了模型预测效果。因此,研究不同预测周期和燃料调节因子分辨率组合对降低林火蔓延预测模型误差的影响是十分必要的。
为降低林火蔓延预测模型的误差,并研究不同预测周期和燃料调节因子分辨率组合对降低林火蔓延预测模型误差的影响。本文基于FARSITE林火蔓延预测模型,提出火线位置-燃料参数双估计方法。该双估计方法利用连续两次观测到的火线位置,通过ETKF和MLP算法对模型的火线位置状态和燃料调节因子参数同时进行估计,进而减少模型状态和模型参数的不确定性。最后,本文研究了不同预测周期和燃料调节因子分辨率对双估计方法降低林火蔓延预测模型误差效果的影响。
本节将对使用的两个关键算法进行介绍。
1)ETKF算法[12]。对于非线性时间离散预测观测系统,ETKF的方程如式(1)、式(2)所示。
xi+1T=MxiT+η
yt0=GxtT+ε
式中:xtT为t时刻m维的“真实状态”;M为非线性预测算子,在这里指FARSITE模型;η为模型误差;yt0为k维观测向量;G是将火线位置从模型空间映射到观测空间的非线性观测算子;ε为模型误差和观测误差。
假设集合数目的大小为N,将初始值x0a进行扰动,生成集合x0a,j,j=1,…,N,预测集合定义如式(3)所示。
xt,jf=Mxt-1,ja
预测平均为xtf的集合平均值。将预测误差协方差矩阵定义为式(4)。
Ptf=1N-1∑j=1Nxt,jf-xtfxt,jf-xtfT
分析集合的计算如下:
使用预测集合矩阵减去预测几何平均值,得到预测集合扰动Ztf0,如式(5)所示。
Ztf0=1N-1xt,1f-xtf,xt,2f-xtf,…,xt,Nf-xtf
为防止滤波器发散并解决背景误差协方差被低估的问题,使用加法形式的误差对预测集合扰动矩阵进行膨胀,如式(6)所示。
Ztf=Ztf0+ζ
式中:ζ为模型误差。
引入一个转换矩阵T将预测集合转换为分析集合,在t时刻的变换矩阵Tt如式(7)所示。
Tt=(I+(HZtf)TR-1HZtf)-1
式中:I为单位矩阵。
分析集合扰动如式(8)所示。
Zta=ZtfTt
随后,分析集合平均值可通过式(9)计算。
xta=xtf+PtfHTHPtfHT+R-1(yt0-G(xtf))
分析集合和分析误差协方差矩阵分别可以定义为式(10)和式(11)。
xt,ja=N-1Zta+xta
Pta=ZtaZtaT
2)MLP算法。MLP中神经元之间的计算和误差传播过程,以及权重和偏置项的更新过程如下[28]:
输入层到隐藏层的加权输入计算公式如式(12)所示。
zj=∑i=1nwijxi+bj
式中:zj为隐藏层神经元j的加权输入;wij为输入层神经元i到隐藏层神经元j的连接权重;xi为输入层神经元i的输入;bj为隐藏层神经元j的偏置项。
隐藏层神经元的输出计算公式如式(13)所示。
aj=f(zj)
式中:aj为隐藏层神经元j的输出;f(*)为激活函数。
隐藏层到输出层的加权输入可由式(14)计算得到。
zk=∑j=1mwjkaj+bk
式中:zk为输出层神经元k的加权输入;wjk为隐藏层神经元j到输出层神经元k的连接权重;aj为隐藏层神经元j的输出;bk为输出层神经元k的偏置项。
输出层神经元的输出计算公式如式(15)所示。
ak=f(zk)
式中:ak为输出层神经元k的输出;f(*)是激活函数。
反向传播算法中的输出层误差项计算公式如式(16)所示。
Ek=yk-ak×f '(zk)
式中:Ek为输出层神经元k的误差项;yk为期望输出;ak为输出层神经元k的输出;f '(*)为激活函数的导数。
反向传播算法中的隐藏层误差项计算如式(17)所示。
Ej=f'zj∑k=1LEjwjk
式中:Ej为隐藏层神经元j的误差项;L为输出层的神经单元数目;f '(*)是激活函数的导数。
权重和偏置项的更新计算公式如式(18)、式(19)所示。
wijnew=wijold+η∂E∂wij
bjnew=bjold+η∂E∂bj
图1展示了森林火灾火场状态-参数双估计方法用于林火蔓延预测的流程图。算法步骤如下。
1)森林火灾发生时,将火灾的初始火线位置及默认燃料调节因子输入FARSITE中,得到t1时刻的预测火线。结合t1时刻的观测火线,利用数据同化ETKF算法得到t1时刻的分析火线。
2)将t1时刻的分析火线及默认燃料调节因子输入FARSITE中,预测下一时刻的火线位置,即t2时刻的预测火线,再结合t2时刻的观测火线,利用数据同化ETKF算法得到t2时刻的分析火线。
3)经过t1和t2时刻连续两次的观测,将t1时刻的观测火线以及随机取样的一组燃料调节因子输入FARSITE中,预测t2时刻的火线位置,即得到t2时刻不同燃料调节因子的预测火线。再将t2时刻不同燃料调节因子的预测火线与t2时刻的分析火线进行对比,计算相似性指数(Similarity Index,SI)。从所有可能的燃料调节因子组合中随机采样625个组合,按照8:2的比例划分为训练集和测试集。使用训练集中的燃料调节因子不断重复步骤3的过程,用于训练MLP神经网络的输入(燃料调节因子)和输出(相似性指数)。
4)利用步骤3得到的燃料调节因子训练集及其对应的相似性指数SI,对MLP神经网络进行训练。训练后可实现对反复运行FARSITE这一过程的替代,从而减少整个双估计的预测时间。例如,当燃料调节因子分辨率为25 m时,共有58=390 625个燃料调节因子组合,即需要运算390 625次FARSITE,才能覆盖可能的燃料调节因子排列组合,而引入MLP后只需运算625次。将燃料调节因子测试集输入训练好的MLP网络,预测得到测试集相似性指数。
5)从训练集相似性指数和测试集相似性指数中选取最大的相似性指数,然后找到最大相似性指数对应的燃料调节因子,即最优燃料调节因子。将步骤2得到的t2时刻分析火线与最优燃料调节因子输入FARSITE中,预测下一时刻的火线位置,即t3时刻的预测火线,亦即利用状态-参数双估计方法预测得到的火线位置。
从图1中可以看出,在第一次和第二次观测的时间间隔内,燃料调节因子取默认值,这是因为燃料调节因子的估计是从第二次观测后开始的,即基于连续两次观测数据估计预测得到的最优燃料调节因子,用于预测下一时刻的火线位置。例如,从t1时刻向后推进到t3时刻的预测火线,最优燃料调节因子是根据t1时刻观测火线和t2时刻观测火线之间的数据估计得出的;再将最优燃料调节因子应用于t3时刻火线位置的预测,以此类推,直至完成所有时刻火线位置的预测。此外,本文建立的算法目前仅适用于不含突变介质场景的预测,因此,t1时刻和t2时刻与预测位置处(t3时刻)的燃料参数具有较好的稳定性。
在林火蔓延预测中,通常需要提供一些方法来比较观测过火区域(或火线)与预测过火区域(或火线)之间的差距,从而为衡量预测方法的性能提供判别指标,主要有两种方法。一种方法是计算预测火线与观测火线之间的距离偏差,如常见的均方根误差(Root Mean Square Error,RMSE),但在计算过程中,RMSE会对观测火线和预测火线中的每个火点进行配对,当观测火点数量和预测火点数量不相等时,会导致RMSE无法准确反映预测方法的整体性能[29]。而另一种方法是比较观测过火区域与预测过火区域之间的形状一致性程度。例如,文献[17]和CHEETHAM A H等[30]提出了一种来源于图像分析领域的Otsuka-Ochiai相似性指数,其能够从整体上评估预测过火区域与观测过火区域之间的形状与位置偏差。Otsuka-Ochiai相似性指数定义为预测过火区域和观测过火区域的相交区域的面积除以它们面积的几何平均值,如式(20)所示。
SI=Aob⋂AfAob×Af
式中:Aob表示观测过火区域面积;Af表示预测过火区域面积;SI表示Otsuka-Ochiai相似性指数值。
Otsuka-Ochiai相似性指数SI的范围在0到1之间,当相似性指数SI=0时,表示预测过火区域与观测过火区域的火线形状和位置完全不一致,反之,当SI=1时表示完全一致。因此,在0至1之间,相似性指数SI的值越大,预测结果与观测数据的相似程度越高,表示火蔓延预测方法的预测精度更高、性能更好;反之亦然。
对于以集合为计算基础的数据同化方法,集合大小直接影响了方法的预测精度和计算成本。理论上,集合大小的增加会提高算法整体的预测精度,但同时也会增加整个过程的计算成本。为合理选择双估计算法的集合大小,以燃料分辨率为50 m、预测周期为4 min的试验工况为例,选择了5组不同的集合大小(N=5、10、20、30、50),通过比较预测结果与真实值之间的均方根误差值来选取合适的集合数目。图2给出了不同的集合大小下预测结果与真实值之间的均方根误差值。
由图2可知,随着集合大小的增加,同一时刻下双估计算法的预测结果与真实值之间的均方根误差值呈减小趋势。当集合大小为20时,所得均方根误差较小,而集合大小进一步增加(N=30、50)时,均方根误差的变化并不明显,但计算成本会进一步增加,因此,在本文双估计算法中数据同化的集合大小选择N=20。
为验证本文所提出的双估计算法,笔者在小兴安岭庆安县曙光林场开展了万平方米尺度的计划烧除试验。此次试验场地为200 m×200 m的矩形地块(点4、6、7和9所围,见图3(a)),研究区域的地形为平坦地形,因此将坡度和坡向设置为0,地形分辨率设置为1 m。
为保证计划烧除试验的可控性及减少蔓延的不确定性,结合专业计划烧除人员的现场经验和火场实际条件,选取图3(a)中左下角点7处作为坐标原点(0,0)进行点火,采用双光无人机(见图3(b))进行火线同步提取。此次计划烧除试验燃烧持续时间为2024年3月24日11:08-11:25,具体如图4所示。
因试验场地燃料类型属性接近FARSITE中的高草燃料模型,因此活植被燃料含水率及1、10、100 h死植被含水率均采用FARSITE默认值,分别为70%、3%、4%和5%。气象数据采用当日气象卫星数据,温度为4 ℃,相对湿度为64%,降雨量为0。采用4台风速仪(沿着点6至点7连线均匀布置,见图 3(a))同步测量风速数据,最后结果取平均值,具体数据见表1。
为分析燃料调节因子分辨率对该双估计方法降低预测误差效果的影响,根据实际林火行为和火场特征(燃料分布、风向),将试验场地沿火势发展方向划分为若干个长条形区域,并对其编号。长条形区域的不同数量代表不同燃料调节因子分辨率,如图5所示。如,将研究区域划分为2个燃料区域,见图5(b),此时燃料调节因子分辨率为100 m,区域编号为31和32,其他情况依此类推。
燃料调节因子估计值从0.6、3.0、5.0、7.0、9.0五个数值中随机组合生成,使每个燃料区域都有5种不同的取值可能。例如,将试验场地划分为8个燃料区域(燃料调节因子分辨率为25 m),共有58=390 625种燃料调节因子组合。作为对照组,仅进行火线位置状态估计,燃料调节因子均采用默认值1。为充分展现双估计算法的预测优势,基于计划烧除试验总时长,将预测周期设置为2、3、4 min,即每隔2、3、4 min对火线位置和燃料调节因子进行1次估计。具体工况设计如表2所示。
为避免人工点火引起的试验误差,选取火势蔓延稳定的11:09时刻观测火线作为起始火线进行预测。图6给出了燃料调节因子分辨率为200 m、预测周期为2 min的试验结果。由图6可知,双估计方法的相似性指数SI值整体上高于单估计方法,即双估计方法预测的火线位置更接近实际的观测火线。但注意到,在某些时刻单估计方法预测的相似性指数SI值会大于双估计方法,例如,在11:15和11:21时刻,单估计方法预测的相似性指数SI值大于双估计方法。
将预测周期分别为2、3、4 min时的相似性指数SI分别取平均值,得到不同预测周期下的单、双估计平均相似性指数SI对比图,见图7。由图7可得:
1)在任何预测周期内,双估计方法的平均相似性指数SI均高于单估计方法,表明双估计方法在火线位置预测方面比单估计方法具有更好的性能。
2)随着预测周期的延长,模型状态误差和参数误差不断累积,单、双估计方法的相似性指数SI值均有所下降,但双估计方法的相似性指数SI值仍显著高于单估计方法。如图7中红色折线所示,随着预测周期增大,双估计方法相比单估计方法的预测精度提升幅度也随之增大。预测周期为2、3、4 min时,双估计方法相比单估计方法的预测精度分别提升了2%、4%、10%左右。
不同燃料调节因子分辨率下,双估计方法的相似性指数SI值均大于单估计方法。将燃料调节因子分辨率为200、100、50、25 m时的相似性指数SI分别取平均值,得到不同燃料调节因子分辨率下的单、双估计平均相似性指数SI对比图,见图8。
由图8可知:
1)任意燃料调节因子分辨率下,双估计方法的平均相似性指数SI值均大于单估计方法的平均相似性指数SI,表明双估计方法较单估计方法在火线位置预测上具有更优越的性能。
2)随着燃料调节因子分辨率的改变,双估计方法相比单估计方法的预测精度提升效果基本稳定在5%左右,表明燃料调节因子分辨率对双估计方法提升预测精度的影响较小。
由3.1节可知,不同预测周期下,双估计方法预测效果整体优于单估计方法,但在个别时刻会出现单估计方法预测的相似性指数SI值大于双估计方法的现象。例如,当燃料调节因子分辨率为200 m、预测周期为2 min时,在11:21时刻单估计方法预测的相似性指数SI值为0.967 6,双估计方法预测的相似性指数SI值为0.939 5。图9(c)中给出了这一过程中单、双估计方法燃料调节因子值的变化情况。由图可知,在11:19时刻,单、双估计方法两次预测所采用的燃料调节因子值差别较大。这是由于本文提出的双估计方法所得的燃料调节因子值是基于该时刻前连续两次观测火线数据估计得到的。由图9(a)可知,从11:17-11:19时刻,观测火线蔓延幅度较大,这会造成双估计算法对燃料调节因子的估计偏大;而未进行燃料调节因子估计的单估计方法,则未受前两个时刻火线蔓延幅度的影响。由于随后的预测周期内(11:19-11:21)火线蔓延幅度实际较小,因此,该时刻参数未更新的单估计方法反而获得了更优的预测效果。类似现象在11:15时刻同样存在(见图9(b)),进一步验证了该结论的普遍性。
图10给出了不同预测周期下实际的观测火线位置图。由图10可知,随着预测周期的延长,上一预测周期内火线蔓延幅度与下一预测周期内火线蔓延幅度逐渐接近,使得双估计方法估计的燃料调节因子和火线位置更加精确。这就能够解释前文所述的预测周期越长,双估计方法预测火线位置的精度相比单估计方法提升越大的原因。
由于实际发生的火灾燃烧每时每刻都在进行,火线位置也在不断地移动变化。因此,结合图10可知,选取一定的预测周期就相当于将火线位置进行“切分”,即人为地将连续的火线位置划分为离散的火线位置。根据前文分析可知,两个相邻预测周期内火线蔓延幅度的一致度对火线位置的预测具有重要影响,而燃料调节因子分辨率仅仅是对燃料信息精度方面的刻画,这在双估计方法执行过程中并未涉及。因此,相较于预测周期,燃料调节因子分辨率对双估计方法降低火蔓延预测模型误差的影响不大。
上文开展的计划烧除试验研究的地形较为平坦,且点火方式为边缘点火,一定程度上减小了蔓延的不确定性。为了进一步验证本文算法在实际场景下的预测性能,收集了2024年4月云南某地发生的一起森林火灾的相关数据,开展案例研究。此次森林火灾陆续投入2 000多名消防救援人员,救援力量以地面力量为主,后期逐步加入航空救援力量,地空配合作战初显成效。火场植被以云南松与阔叶林混交为主,油性大、燃烧急、火势猛、速度快,在持续强风作用下,呈多种形态向四周扩散蔓延。打火时间多集中在夜间,原因是上午10时到晚上7时风力大、温度高,不宜开展作战,每天实际作战时间约为4~5 h。经过5天4夜的扑救,16日9时左右,灭火作战取得胜利,最终过火面积约为24 km2。
基于地理空间数据云获取该地区的数字高程数据(分辨率12.5 m),利用ArcGIS等地理数据处理工具,提取高程数据中的海拔、坡度、坡向等地形特征信息,具体如图11所示。此外,利用公开气象网站获取火场周边区域起火后一周的温度、湿度、降雨量、风速、风向等气象信息,每小时更新1次,详细数据如图12所示。
基于该起历史火灾案例,开展了本文算法的性能验证。结合地形、气象和植被等输入数据,基于数据同化方法的双估计算法,在预测过程中不断融入观测到的火线位置,得到下一时刻的火线位置预测结果,并将其与下一时刻实际观测火线位置进行对比。结果如图13所示。
豪斯多夫距离(Hausdorff Distance)是一种用于测量两个集合之间距离的方法。这种距离度量特别适用于比较形状和轮廓的相似性,在图像处理和计算机视觉等领域广泛应用。豪斯多夫距离通过衡量一个集合中每个点到另一个集合中最近点距离的最大值来评估两个集合的相似程度。在本文中,豪斯多夫距离可用于表征预测结果与真实火线位置误差最大处的偏差,而相似性指数则计算了预测结果与真实火线位置的整体轮廓相似程度。从相似性指数的定量结果看出,本文算法的火线位置预测结果整体表现较好。从豪斯多夫距离的计算结果看出,本文算法在部分位置处仍存在一定误差。其误差来源主要来源于气象、植被等输入参数的不确定性和消防力量的干预。总之,数据驱动方法通过融入观测数据校正了预测火线位置,达到了提升预测精度的目的。该案例研究结果展现了数据驱动方法在实战应用中的潜力。
针对林火蔓延预测模型中火线状态和燃料参数的不确定性,本文基于连续两次观测火线位置数据,结合数据同化ETKF算法及MLP神经网络分别对火线位置和输入参数进行实时估计,建立了火场状态-参数双估计算法,并结合万平方米尺度的计划烧除试验和实际历史火灾案例进行了验证,通过试验结果与分析得到以下结论。
1)在不同预测周期和燃料调节因子分辨率下,双估计方法的平均相似性指数SI均高于单估计方法;随着预测周期的增大,单、双估计方法的相似性指数SI值均有所下降,但双估计方法的相似性指数SI值仍显著高于单估计方法。燃料调节因子分辨率对双估计方法提升预测精度的影响较小。
2)在历史火灾案例中,双估计算法的火线位置预测结果整体表现良好,存在的误差主要来源于气象、植被等输入参数的不确定性和消防力量的干预,验证了本文算法在实际场景下的预测性能。

参考文献 引证文献
排序方式:
1
CUNNINGHAM C X,WILLIAMSON G J,BOWMAN D M J S.Increasing frequency and intensity of the most extreme wildfires on Earth[J].Nature Ecology & Evolution,2024,8(8):1420-1425.
2
LINN R R,Harlow F H.Use of transport models for wildfire behavior simulations[R].Las Cruces:Los Alamos National Lab(LANL),1998.
3
MELL W,MARANGHIDES A,MCDERMOTT R,et al.Numerical simulation and experiments of burning Douglas fir trees[J].Combustion and Flame,2009,156(10):2023-2041.
4
MELL W,JENKINS M A,GOULD J,et al.A physics-based approach to modelling grassland fires[J].International Journal of Wildland Fire,2007,16(1):1-22.
5
王正非.山火初始蔓延速度测算法[J].山地研究,1983,1(2):42-51.
6
张晓婷,刘培顺,王学芳.王正非林火蔓延模型改进研究[J].山东林业科技,2020,50(1):1-6,40.
7
王晓红,王千雪,于宏影,等.FARSITE模型原理、结构及其在林火行为模拟中的应用[J].温带林业研究,2022,5(4):55-60.
8
FINNEY M A.FARSITE:Fire area simulator-model develop⁃ ment and evaluation[R].Ogden:USDA Forest Service Rocky⁃ Mountain Research Station,2004.
9
SULLIVAN A L.Wildland surface fire spread modelling,1990-2007.1:Physical and quasi-physical models[J].International Journal of Wildland Fire,2009,18(4):349-368.
10
钱兰,张启兴,张永明.集合卡曼滤波算法对FARSITE模型林火蔓延预测的修正效果研究[J].火灾科学,2020,29(1):32-41.
11
ZHOU T J,JI J,JIANG Y,et al.EnKF-based real-time prediction of wildfire propagation[C]//The Proceedings of 11th Asia-Oceania Symposium on Fire Science and Technology.Singapore:Springer Singapore,2020:713-724.
12
ZHOU T J,DING L,JI J,et al.Ensemble transform Kalman filter (ETKF) for large-scale wildland fire spread simulation using FARSITE tool and state estimation method[J].Fire Safety Journal,2019,105:95-106.
13
唐纯.无人机实时动态数据驱动的林火蔓延模拟优化方法研究[D].哈尔滨:东北林业大学,2023.
14
王丹.林火蔓延中的数据同化方法研究[D].长沙:中南林业科技大学,2017.
15
胡全.基于GIS的森林火场模拟关键技术研究[D].哈尔滨:东北林业大学,2014.
16
汪瑾.基于长短期记忆网络的林火风险与蔓延预测研究[D].南京:南京林业大学,2023.
17
ZHOU T J,DING L,JI J,et al.Combined estimation of fire perimeters and fuel adjustment factors in FARSITE for forecasting wildland fire propagation[J].Fire Safety Journal,2020,116:103167.
18
PRASAD D V ,SARASWATHI S.The role of anomaly detection in industry 4.0:A survey of techniques and applications[J].Journal of Trends in Computer Science and Smart Technology,2024,6(2):125-138.
19
谢洋,陆新建,邵晓雯,等.基于MLP-RF组合模型预测加氢裂化装置柴油产量[J].化学工程,2024,52(11):66-70.
20
何卫,牛岱松,宋世琦.实测数据驱动的地铁振动源强MLP神经网络预测方法[J/OL].振动工程学报,1-9.
21
黄德启,涂亚婷,张振华,等.基于MLP与改进GCN-TD3的交通信号控制建模与仿真[J].系统仿真学报,2025,37(10):2568-2577.
22
武慧荣,陈少阳,崔淑华.基于GM(1,1)-MLP神经网络模型的大宗货物运输需求预测[J].公路交通科技,2023,40(10):233-240.
23
茅晓晨,焦英辰,陈璇,等.基于POD和多层感知器的跨声速压气机叶型精细优化设计方法[J].海军航空大学学报,2024,39(6):715-725.
24
窦淼,侯祥宁.PCA-MLP神经网络模型在黄河宁夏段径流预测中的应用[J].水利信息化,2024(4):49-53.
25
肖舜祯,刘强,徐志扬,等.基于MLP的上海市主要树种单木胸径生长率模型[J].江西农业大学学报,2022,44(5):1169-1176.
26
TANEJA R,HILTON J,WALLACE L,et al.Effect of fuel spatial resolution on predictive wildfire models[J].International Journal of Wildland Fire,2021,30(10):776-789.
27
张时雨.基于机载监测系统的东北森林地表火蔓延与预测模型研究[D].哈尔滨:东北林业大学,2022.
28
胡宏宇.基于MLP的海上无人跨域协同效能评估系统研究与实现[D].沈阳:辽宁大学,2024.
29
YINKA-BANJO C O,AKINYEMI M I,YAMA B B.Ensemble algorithm for simulated corrosion data-tentative[J].Annals of Science and Technology,2023,8(2):56-62.
30
CHEETHAM A H,HAZEL J E.Binary (presence-absence) similarity coefficients[J].Journal of Paleontology,1969,43(5):1130-1136.
2026年第45卷第5期
PDF下载
6
2
引用本文
BibTeX
文章信息
doi: 10.20168/j.1009-0029.2026.05.0132.11
  • 接收时间:2025-02-16
  • 首发时间:2026-10-22
  • 出版时间:2026-05-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-02-16
  • 修回日期:2025-05-02
基金
作者信息
    1中国科学技术大学 火灾安全全国重点实验室,安徽 合肥 230026
    2中国科学技术大学 先进技术研究院,安徽 合肥 230031

通讯作者:

纪杰,中国科学技术大学研究员,博士生导师,。
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/xfkxyjs/CN/10.20168/j.1009-0029.2026.05.0132.11
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏