Article(id=1251559135813255893, tenantId=1146029695717560320, journalId=1146031591421210625, issueId=1251559134307500754, articleNumber=null, orderNo=null, doi=10.3981/j.issn.1000-7857.2025.12.00116, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1765296000000, receivedDateStr=2025-12-10, revisedDate=1772208000000, revisedDateStr=2026-02-28, acceptedDate=null, acceptedDateStr=null, onlineDate=1776324550822, onlineDateStr=2026-04-16, pubDate=1774627200000, pubDateStr=2026-03-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1776324550822, onlineIssueDateStr=2026-04-16, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1776324550822, creator=13701087609, updateTime=1776324550822, updator=13701087609, issue=Issue{id=1251559134307500754, tenantId=1146029695717560320, journalId=1146031591421210625, year='2026', volume='44', issue='6', pageStart='1', pageEnd='112', issueExtLink='null', onlineDate='null', pubDate='1774627200000', pubDateStr='2026-03-28', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1776324550463, creator='13701087609', updateTime=1776332259412, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1251591468125733729, tenantId=1146029695717560320, journalId=1146031591421210625, issueId=1251559134307500754, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1251591468125733730, tenantId=1146029695717560320, journalId=1146031591421210625, issueId=1251559134307500754, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=48, endPage=56, ext={EN=ArticleExt(id=1251559136262046426, articleId=1251559135813255893, tenantId=1146029695717560320, journalId=1146031591421210625, language=EN, title=Noisy multivariate prediction via filtering and multi−scale temporal synergy, columnId=1150494642224591153, journalTitle=Science & Technology Review, columnName=Exclusive, runingTitle=null, highlight=null, articleAbstract=

Noisy multivariate prediction is a common challenge in fields such as environmental science, transportation, and industry. The core difficulty lies in balancing noise filtering with multi−scale feature capture. To address this, a hybrid model (Kalman−LSTM−Transformer) based on Kalman filter, long short−term memory (LSTM), and Transformer is proposed. This model captures local temporal and global dependencies while filtering noise, and integrates Bayesian optimization to achieve efficient and accurate prediction. Using open−pit mine dust concentration prediction as a case study, experiments based on 1−year of monitoring data demonstrate that the model outperforms baseline models, reducing the root mean square error (RMSE) by 21.70%–27.19% and the mean absolute error (MAE) by 6.68%–18.30%, while achieving a coefficient of determination (R2) of 0.934. Ablation experiments and hyperparameter analysis results further confirm the effectiveness of each module. The model exhibits transferability to similar scenarios, providing support for intelligent early warning and precision management across multiple domains.

, authors=null, authorsList=Qiwen GUO, Yulong ZHANG, Xin ZHANG, Dejie LI, Xin WANG, Haixia ZHANG, Yuhan WANG, authorCompany=null, correspAuthors=Yulong ZHANG, authorNote=null, correspAuthorsNote=null, copyrightStatement=All rights reserved. Unauthorized reproduction is prohibited., 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=1251559138036237073, articleId=1251559135813255893, tenantId=1146029695717560320, journalId=1146031591421210625, language=CN, title=基于滤波与多尺度时序协同的含噪多变量预测, columnId=1150494642375586098, journalTitle=科技导报, columnName=特色专题, runingTitle=null, highlight=null, articleAbstract=

含噪多变量预测是环境、交通、工业等领域的共性难题,其核心挑战在于平衡噪声过滤与多尺度特征捕捉。提出一种基于卡尔曼滤波−长短期记忆网络−Transformer的混合模型(Kalman−LSTM−Transformer)。该模型在过滤噪声的同时捕捉局部时序与全局依赖,并结合贝叶斯优化以实现高效精准预测。以露天矿粉尘浓度预测为案例验证,基于1年监测数据的实验显示,该模型较基准模型的均方根误差降低21.70%~27.19%,平均绝对误差降低6.68%~18.30%,决定系数R2达0.934;消融实验与超参数分析结果进一步证实了各模块有效性。该模型可迁移至同类场景,为多领域智能预警与精准治理提供支撑。

, authors=

郭启文,博士研究生,研究方向为露天矿粉尘防控,电子信箱:

, authorsList=郭启文, 张玉龙, 张欣, 李德杰, 汪昕, 张海霞, 王宇涵, authorCompany=null, correspAuthors=张玉龙, authorNote=null, correspAuthorsNote=
张玉龙(通信作者),副教授,研究方向为多场耦合岩体力学,电子信箱:
, copyrightStatement=版权所有,未经授权,不得转载。, copyrightOwner=《科技导报》编辑部, extLink=null, articleAbsUrl=null, sourceXml=NpYYbweMRuoSbb0D84cGgw==, magXml=kD9T8eSDPjCFRUFU4UuKng==, pdfUrl=null, pdf=d2B6k+x3QfYcdAF5UMfUNw==, pdfFileSize=2586252, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=a+ROIb880Iqe2RH/IBsPHA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=Z+GUW/N2jrowOirb6I/4TQ==, mapNumber=null, fund=null)}, authors=[Author(id=1251567849513365577, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2410425@stu.neu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1251567849572085835, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849513365577, language=EN, stringName=Qiwen GUO, firstName=Qiwen, middleName=null, lastName=GUO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567849639194700, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849513365577, language=CN, stringName=郭启文, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio={"content":"

郭启文,博士研究生,研究方向为露天矿粉尘防控,电子信箱:

"}, bioImg=null, bioContent=

郭启文,博士研究生,研究方向为露天矿粉尘防控,电子信箱:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567849693720654, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhangyulong@mail.neu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1251567849752440912, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849693720654, language=EN, stringName=Yulong ZHANG, firstName=Yulong, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567849836326993, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849693720654, language=CN, stringName=张玉龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567849890852947, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, 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=1251567849966350421, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849890852947, language=EN, stringName=Xin ZHANG, firstName=Xin, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2College of Civil Engineering, Shandong Jianzhu University, Jinan 250101, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850025070678, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849890852947, language=CN, stringName=张欣, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2山东建筑大学土木工程学院,济南 250101, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567849190404162, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=2, ext=[AuthorCompanyExt(id=1251567849202987075, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849190404162, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2College of Civil Engineering, Shandong Jianzhu University, Jinan 250101, China), AuthorCompanyExt(id=1251567849211375684, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849190404162, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2山东建筑大学土木工程学院,济南 250101)])]), Author(id=1251567850079596632, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=3, 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=1251567850150899802, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850079596632, language=EN, stringName=Dejie LI, firstName=Dejie, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3CCCC Second Harbour Engineering Co., Ltd., Wuhan 430040, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850209620059, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850079596632, language=CN, stringName=李德杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3中交第二航务工程局有限公司,武汉 430040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567849282678853, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=3, ext=[AuthorCompanyExt(id=1251567849320427590, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849282678853, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3CCCC Second Harbour Engineering Co., Ltd., Wuhan 430040, China), AuthorCompanyExt(id=1251567849324621895, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849282678853, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3中交第二航务工程局有限公司,武汉 430040)])]), Author(id=1251567850268340317, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, 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=1251567850335449183, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850268340317, language=EN, stringName=Xin WANG, firstName=Xin, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850406752352, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850268340317, language=CN, stringName=汪昕, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567850524192866, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=5, 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=1251567850603884644, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850524192866, language=EN, stringName=Haixia ZHANG, firstName=Haixia, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850666799205, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850524192866, language=CN, stringName=张海霞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567850721325159, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=6, 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=1251567850784239721, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850721325159, language=EN, stringName=Yuhan WANG, firstName=Yuhan, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850876514410, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850721325159, language=CN, stringName=王宇涵, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])])], keywords=[Keyword(id=1251567851031703659, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, orderNo=1, keyword=noisy multivariate prediction), Keyword(id=1251567851103006828, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, orderNo=2, keyword=Kalman filter), Keyword(id=1251567851157532781, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, orderNo=3, keyword=long short−term memory), Keyword(id=1251567851245613166, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, orderNo=4, keyword=Transformer), Keyword(id=1251567851329499247, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, orderNo=5, keyword=Bayesian optimization), Keyword(id=1251567851409191024, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, orderNo=1, keyword=含噪多变量预测), Keyword(id=1251567851484688497, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, orderNo=2, keyword=卡尔曼滤波), Keyword(id=1251567851555991666, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, orderNo=3, keyword=长短期记忆网络), Keyword(id=1251567851623100531, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, orderNo=4, keyword=Transformer), Keyword(id=1251567851690209396, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, orderNo=5, keyword=贝叶斯优化)], refs=[Reference(id=1251567855737712828, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=42, issue=20, pageStart=23, pageEnd=31, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=毛玉如, 汪赟, 张建勋, journalName=科技导报, refType=null, unstructuredReference=毛玉如, 汪赟, 张建勋. 中国噪声自动监测的现状和发展[J]. 科技导报, 2024, 42(20): 23-31., articleTitle=中国噪声自动监测的现状和发展, refAbstract=null), Reference(id=1251567855855153346, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=42, issue=20, pageStart=85, pageEnd=94, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=袁旻忞, 王彦琴, 邵社刚, journalName=科技导报, refType=null, unstructuredReference=袁旻忞, 王彦琴, 邵社刚, . 道路交通噪声控制技术研发进展[J]. 科技导报, 2024, 42(20): 85-94., articleTitle=道路交通噪声控制技术研发进展, refAbstract=null), Reference(id=1251567855951622343, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=42, issue=20, pageStart=70, pageEnd=84, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=户文成, 刘磊, 康钟绪, journalName=科技导报, refType=null, unstructuredReference=户文成, 刘磊, 康钟绪, . 城市轨道交通噪声污染防治进展[J]. 科技导报, 2024, 42(20): 70-84., articleTitle=城市轨道交通噪声污染防治进展, refAbstract=null), Reference(id=1251567856056479951, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2019, volume=5, issue=3, pageStart=67, pageEnd=83, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=Entwistle J A, Hursthouse A S, Marinho Reis P A, journalName=Current Pollution Reports, refType=null, unstructuredReference=Entwistle J A, Hursthouse A S, Marinho Reis P A, et al. Metalliferous mine dust: Human health impacts and the potential determinants of disease in mining communities[J]. Current Pollution Reports, 2019, 5(3): 67-83., articleTitle=Metalliferous mine dust: Human health impacts and the potential determinants of disease in mining communities, refAbstract=null), Reference(id=1251567856249417940, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2023, volume=25, issue=6, pageStart=1903, pageEnd=1920, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=Hosseini S, Poormirzaee R, Gilani S O, journalName=Clean Technologies and Environmental Policy, refType=null, unstructuredReference=Hosseini S, Poormirzaee R, Gilani S O, et al. A reliability−based rock engineering system for clean blasting: Risk analysis and dust emissions forecasting[J]. Clean Technologies and Environmental Policy, 2023, 25(6): 1903-1920., articleTitle=A reliability−based rock engineering system for clean blasting: Risk analysis and dust emissions forecasting, refAbstract=null), Reference(id=1251567856396218585, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2020, volume=17, issue=5, pageStart=1570, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=Hendryx M, Islam M S, Dong G H, journalName=International Journal of Environmental Research and Public Health, refType=null, unstructuredReference=Hendryx M, Islam M S, Dong G H, et al. Air pollution emissions 2008−2018 from Australian coal mining: Implications for public and occupational health[J]. International Journal of Environmental Research and Public Health, 2020, 17(5): 1570., articleTitle=Air pollution emissions 2008−2018 from Australian coal mining: Implications for public and occupational health, refAbstract=null), Reference(id=1251567856492687581, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2019, volume=78, issue=10, pageStart=310, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=Guo X D, Sun Q F, Zhao Y S, journalName=Environmental Earth Sciences, refType=null, unstructuredReference=Guo X D, Sun Q F, Zhao Y S. et al. Identification and characterisation of heavy metals in farmland soil of Hunchun basin[J]. Environmental Earth Sciences, 2019, 78(10): 310., articleTitle=et al. Identification and characterisation of heavy metals in farmland soil of Hunchun basin, refAbstract=null), Reference(id=1251567858078134499, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=5574579, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=Zhang H, Han W H, Xu Y L, journalName=Journal of Healthcare Engineering, refType=null, unstructuredReference=Zhang H, Han W H, Xu Y L, et al. Analysis on the development status of coal mine dust disaster prevention technology in China[J]. Journal of Healthcare Engineering, 2021: 5574579., articleTitle=Analysis on the development status of coal mine dust disaster prevention technology in China, refAbstract=null), Reference(id=1251567858145243365, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=15, issue=11, pageStart=102273, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=Li X, Zhao H, Cheng J Y, journalName=Atmospheric Pollution Research, refType=null, unstructuredReference=Li X, Zhao H, Cheng J Y, et al. BWO−BiLSTM & CNN composite model for prediction of atmospheric particulate matter mass concentration[J]. Atmospheric Pollution Research, 2024, 15(11): 102273., articleTitle=BWO−BiLSTM & CNN composite model for prediction of atmospheric particulate matter mass concentration, refAbstract=null), Reference(id=1251567858250100970, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2021, volume=51, issue=4, pageStart=2534, pageEnd=2555, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=Chen Y C, Li D C, journalName=Applied Intelligence, refType=null, unstructuredReference=Chen Y C, Li D C. Selection of key features for PM2.5 prediction using a wavelet model and RBF−LSTM[J]. Applied Intelligence, 2021, 51(4): 2534-2555., articleTitle=Selection of key features for PM2.5 prediction using a wavelet model and RBF−LSTM, refAbstract=null), Reference(id=1251567858342375661, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=22, issue=null, pageStart=102305, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=Yadav V, Yadav A K, Singh V, journalName=Results in Engineering, refType=null, unstructuredReference=Yadav V, Yadav A K, Singh V, et al. Artificial neural network an innovative approach in air pollutant prediction for environmental applications: A review[J]. Results in Engineering, 2024, 22: 102305., articleTitle=Artificial neural network an innovative approach in air pollutant prediction for environmental applications: A review, refAbstract=null), Reference(id=1251567858426261742, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2025, volume=19, issue=3, pageStart=774, pageEnd=786, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=温雯, 胡智鑫, 郝志峰, journalName=计算机科学与探索, refType=null, unstructuredReference=温雯, 胡智鑫, 郝志峰. 面向序列推荐的深度化指数移动平均学习方法[J]. 计算机科学与探索, 2025, 19(3): 774-786., articleTitle=面向序列推荐的深度化指数移动平均学习方法, refAbstract=null), Reference(id=1251567858510147825, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2020, volume=8, issue=3, pageStart=103208, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=Leong W C, Kelani R O, Ahmad Z, journalName=Journal of Environmental Chemical Engineering, refType=null, unstructuredReference=Leong W C, Kelani R O, Ahmad Z. Prediction of air pollution index (API) using support vector machine (SVM)[J]. Journal of Environmental Chemical Engineering, 2020, 8(3): 103208., articleTitle=Prediction of air pollution index (API) using support vector machine (SVM), refAbstract=null), Reference(id=1251567858623394038, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=133, issue=null, pageStart=108399, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=Kamran Kamran, Jiskani I M, Wang Z, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=Kamran Kamran, Jiskani I M, Wang Z, et al. Decision intelligence−driven predictive modelling of air quality index in surface mining[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 108399., articleTitle=Decision intelligence−driven predictive modelling of air quality index in surface mining, refAbstract=null), Reference(id=1251567858719863033, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2024, volume=436, issue=null, pageStart=140411, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=Yang Y K, Zhou W, Wang Z M, journalName=Journal of Cleaner Production, refType=null, unstructuredReference=Yang Y K, Zhou W, Wang Z M, et al. Accurate long−term dust concentration prediction in open−pit mines: A novel machine learning approach integrating meteorological conditions and mine production intensity[J]. Journal of Cleaner Production, 2024, 436: 140411., articleTitle=Accurate long−term dust concentration prediction in open−pit mines: A novel machine learning approach integrating meteorological conditions and mine production intensity, refAbstract=null), Reference(id=1251567858807943423, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2019, volume=130, issue=null, pageStart=104909, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=15, authorNames=Di Q, Amini H, Shi L H, journalName=Environment International, refType=null, unstructuredReference=Di Q, Amini H, Shi L H, et al. An ensemble−based model of PM2.5 concentration across the contiguous United States with high spatiotemporal resolution[J]. Environment International, 2019, 130: 104909., articleTitle=An ensemble−based model of PM2.5 concentration across the contiguous United States with high spatiotemporal resolution, refAbstract=null), Reference(id=1251567858887635203, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2020, volume=701, issue=null, pageStart=134474, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=16, authorNames=Choubin B, Abdolshahnejad M, Moradi E, journalName=Science of The Total Environment, refType=null, unstructuredReference=Choubin B, Abdolshahnejad M, Moradi E, et al. Spatial hazard assessment of the PM10 using machine learning models in Barcelona, Spain[J]. Science of The Total Environment, 2020, 701: 134474., articleTitle=Spatial hazard assessment of the PM10 using machine learning models in Barcelona, Spain, refAbstract=null), Reference(id=1251567858988298500, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2021, volume=4, issue=12, pageStart=2100220, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=17, authorNames=Li P, Wang S W, Ji H, journalName=Advanced Theory and Simulations, refType=null, unstructuredReference=Li P, Wang S W, Ji H, et al. Air quality index prediction based on an adaptive dynamic particle swarm optimized bidirectional gated recurrent neural network–China region[J]. Advanced Theory and Simulations, 2021, 4(12): 2100220., articleTitle=Air quality index prediction based on an adaptive dynamic particle swarm optimized bidirectional gated recurrent neural network–China region, refAbstract=null), Reference(id=1251567859059601673, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2022, volume=204, issue=null, pageStart=117600, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=18, authorNames=Liu B C, Yu Z C, Wang Q S, journalName=Expert Systems with Applications, refType=null, unstructuredReference=Liu B C, Yu Z C, Wang Q S, et al. Prediction of SSE Shanghai Enterprises index based on bidirectional LSTM model of air pollutants[J]. Expert Systems with Applications, 2022, 204: 117600., articleTitle=Prediction of SSE Shanghai Enterprises index based on bidirectional LSTM model of air pollutants, refAbstract=null), Reference(id=1251567859172847886, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2022, volume=13, issue=5, pageStart=223, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=19, authorNames=Ding C, Zheng Z Y, Zheng S R, journalName=Information, refType=null, unstructuredReference=Ding C, Zheng Z Y, Zheng S R, et al. Accurate air−quality prediction using genetic−optimized gated−recurrent−unit architecture[J]. Information, 2022, 13(5): 223., articleTitle=Accurate air−quality prediction using genetic−optimized gated−recurrent−unit architecture, refAbstract=null), Reference(id=1251567859269316881, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2025, volume=269, issue=null, pageStart=112463, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=20, authorNames=Zhong J D, Liu J L, journalName=Building and Environment, refType=null, unstructuredReference=Zhong J D, Liu J L. Seq2seq modelling for cross−site temporal forecasting of urban air pollutant concentrations leveraging sensor data[J]. Building and Environment, 2025, 269: 112463., articleTitle=Seq2seq modelling for cross−site temporal forecasting of urban air pollutant concentrations leveraging sensor data, refAbstract=null), Reference(id=1251567859378368793, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2022, volume=15, issue=9, pageStart=1533, pageEnd=1546, url=null, language=null, rfNumber=[22], rfOrder=21, authorNames=Chen Y B, Chen X M, Xu A L, journalName=Air Quality, Atmosphere & Health, refType=null, unstructuredReference=Chen Y B, Chen X M, Xu A L, et al. A hybrid CNN−Transformer model for ozone concentration prediction[J]. Air Quality, Atmosphere & Health, 2022, 15(9): 1533-1546., articleTitle=A hybrid CNN−Transformer model for ozone concentration prediction, refAbstract=null), Reference(id=1251567859470643487, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2025, volume=327, issue=null, pageStart=136192, pageEnd=null, url=null, language=null, rfNumber=[23], rfOrder=22, authorNames=Gong Y, Wang Y Z, Xie Y H, journalName=Energy, refType=null, unstructuredReference=Gong Y, Wang Y Z, Xie Y H, et al. Dynamic fusion LSTM−Transformer for prediction in energy harvesting from human motions[J]. Energy, 2025, 327: 136192., articleTitle=Dynamic fusion LSTM−Transformer for prediction in energy harvesting from human motions, refAbstract=null), Reference(id=1251567859541946660, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, doi=null, pmid=null, pmcid=null, year=2025, volume=178, issue=null, pageStart=113033, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=23, authorNames=Chen D Y, Liu H, journalName=Applied Soft Computing, refType=null, unstructuredReference=Chen D Y, Liu H. Integrating multi−dimensional graph attention networks and transformer architecture for predicting air pollution in subway stations[J]. Applied Soft Computing, 2025, 178: 113033., articleTitle=Integrating multi−dimensional graph attention networks and transformer architecture for predicting air pollution in subway stations, refAbstract=null)], funds=[Fund(id=1251567855444111530, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, awardId=2023YFC2907303, language=CN, fundingSource=国家重点研发计划项目(2023YFC2907303), fundOrder=null, country=null), Fund(id=1251567855532191919, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, awardId=12372374, language=CN, fundingSource=国家自然科学基金面上项目(12372374), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)]), AuthorCompany(id=1251567849190404162, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=2, ext=[AuthorCompanyExt(id=1251567849202987075, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849190404162, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2College of Civil Engineering, Shandong Jianzhu University, Jinan 250101, China), AuthorCompanyExt(id=1251567849211375684, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849190404162, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2山东建筑大学土木工程学院,济南 250101)]), AuthorCompany(id=1251567849282678853, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=3, ext=[AuthorCompanyExt(id=1251567849320427590, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849282678853, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3CCCC Second Harbour Engineering Co., Ltd., Wuhan 430040, China), AuthorCompanyExt(id=1251567849324621895, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849282678853, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3中交第二航务工程局有限公司,武汉 430040)])], figs=[ArticleFig(id=1251567851866370165, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=KaWIy58UFynPjhr0mcCStQ==, figureFileBig=oBMtWg1pbsH7NwqIv0W3nQ==, tableContent=null), ArticleFig(id=1251567851954450550, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=图1, caption=Kalman−LSTM−Transformer模型预测流程, figureFileSmall=KaWIy58UFynPjhr0mcCStQ==, figureFileBig=oBMtWg1pbsH7NwqIv0W3nQ==, tableContent=null), ArticleFig(id=1251567852055113847, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=QMMiolBZD8MLi5lQa2x0PA==, figureFileBig=DgaAOxd0fujuekoKCSdaKw==, tableContent=null), ArticleFig(id=1251567853632172155, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=图2, caption=LSTM模块组成结构, figureFileSmall=QMMiolBZD8MLi5lQa2x0PA==, figureFileBig=DgaAOxd0fujuekoKCSdaKw==, tableContent=null), ArticleFig(id=1251567853728641149, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=2aFLut2v3cuEZTd19JO3Vg==, figureFileBig=TsiEC7JqH7bvafPEoS4Xcw==, tableContent=null), ArticleFig(id=1251567853816721537, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=图3, caption=Transformer模块组成架构, figureFileSmall=2aFLut2v3cuEZTd19JO3Vg==, figureFileBig=TsiEC7JqH7bvafPEoS4Xcw==, tableContent=null), ArticleFig(id=1251567853917384834, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=Z4QVGFZlm2QU4jZnvrEBhw==, figureFileBig=o0EyS9DFVHz0toskSEGsRQ==, tableContent=null), ArticleFig(id=1251567854097739907, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=图4, caption=各种模型在测试集上的预测结果, figureFileSmall=Z4QVGFZlm2QU4jZnvrEBhw==, figureFileBig=o0EyS9DFVHz0toskSEGsRQ==, tableContent=null), ArticleFig(id=1251567854190014596, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=AIFxl4+thRVpeERmqlIang==, figureFileBig=Sxk5/WdxIUoGT+DxInGy0w==, tableContent=null), ArticleFig(id=1251567854252929159, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=图5, caption=消融实验预测相对误差分布, figureFileSmall=AIFxl4+thRVpeERmqlIang==, figureFileBig=Sxk5/WdxIUoGT+DxInGy0w==, tableContent=null), ArticleFig(id=1251567854361981066, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=mKLLRaDjTrSJqkGmipOJkA==, figureFileBig=6zRIHXJUa0bqJ3+rafiIJQ==, tableContent=null), ArticleFig(id=1251567854454255756, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=图6, caption=不同超参数优化策略下预测评价指标结果对比, figureFileSmall=mKLLRaDjTrSJqkGmipOJkA==, figureFileBig=6zRIHXJUa0bqJ3+rafiIJQ==, tableContent=null), ArticleFig(id=1251567854559113359, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
数据类型粉尘浓度/
(μg·m-3
露点温度/
温度/℃气压/kPa风速/(m·s-1雪量/mm雨量/mm
最大值980.0028.0041.00104.3064.3627.0036.00
最小值8.00−28.00−19.0099.400.4500
平均值96.361.6511.68101.6316.540.070.27
中位数71.002.0012.00101.506.7100
标准差92.8814.3812.609.6714.490.981.81
), ArticleFig(id=1251567854672359569, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=表1, caption=

粉尘浓度数据集基本信息

, figureFileSmall=null, figureFileBig=null, tableContent=
数据类型粉尘浓度/
(μg·m-3
露点温度/
温度/℃气压/kPa风速/(m·s-1雪量/mm雨量/mm
最大值980.0028.0041.00104.3064.3627.0036.00
最小值8.00−28.00−19.0099.400.4500
平均值96.361.6511.68101.6316.540.070.27
中位数71.002.0012.00101.506.7100
标准差92.8814.3812.609.6714.490.981.81
), ArticleFig(id=1251567854861103252, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
模型RMSEMAER2
LSTM41.70724.7150.876
Seq2Seq41.75124.9590.874
Transformer38.82121.8510.892
Kalman−LSTM−Transformer30.39820.3910.934
), ArticleFig(id=1251567854978543766, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=表2, caption=

不同模型预测评价指标结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型RMSEMAER2
LSTM41.70724.7150.876
Seq2Seq41.75124.9590.874
Transformer38.82121.8510.892
Kalman−LSTM−Transformer30.39820.3910.934
), ArticleFig(id=1251567855095984285, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
模型RMSEMAER2
LSTM−Transformer36.08221.7150.906
Kalman−LSTM34.12521.0220.918
Kalman−Transformer33.99620.9810.922
Kalman−LSTM−Transformer30.39820.3910.934
), ArticleFig(id=1251567855251173537, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, language=CN, label=表3, caption=

消融实验预测评价指标结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型RMSEMAER2
LSTM−Transformer36.08221.7150.906
Kalman−LSTM34.12521.0220.918
Kalman−Transformer33.99620.9810.922
Kalman−LSTM−Transformer30.39820.3910.934
)], attaches=null, journal=Journal(id=1125356956822126595, delFlag=0, nameCn=科技导报, nameEn=Science & Technology Review, nameHistory1=null, nameHistory2=null, issn=1000-7857, eissn=, cn=11-1421/N, coden=null, periodic=3, language=CN, oaType=0, 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=wfghvu3bhh/dKxuZ+ucVHA==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Rev, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1784015846012, createdBy=null, updatedBy=13041195026, firstLetterCn=K, firstLetterEn=K, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=wfghvu3bhh/dKxuZ+ucVHA==, picEn=yjSfclmpNm7ihn9NbTZ69g==, jcr=null, cjcr=null, exts=[JournalExt(id=1283818766098219763, language=CN, name=科技导报, nameHistory1=null, nameHistory2=null, managedBy=中国科学技术协会, sponsoredBy=中国科学技术协会, publishedBy=科技导报社, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.kjdb.org/CN/home, createdTime=1784015846037, updatedTime=1784015846037, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=http://www.kjdb.org/CN/column/column7.shtml, submissionAuthorUrl=https://kjdbauthor.cast.org.cn/webm, submissionEditorUrl=https://kjdbeditor.cast.org.cn/webm/, submissionReviewUrl=https://kjdbauthor.cast.org.cn/webm, submissionCeEditorUrl=https://kjdbeditor.cast.org.cn/webm/, submissionAeEditorUrl=https://kjdbeditor.cast.org.cn/webm/, option={"copyright":""}), JournalExt(id=1283818766144357108, language=EN, name=Science & Technology Review, 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=http://www.kjdb.org/EN/home, createdTime=1784015846048, updatedTime=1784015846048, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=http://www.kjdb.org/EN/column/column7.shtml, submissionAuthorUrl=https://kjdbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://kjdbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://kjdbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://kjdbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://kjdbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031591421210625, websiteList=[Website(id=1146104741081231361, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031591421210625, 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/kjdb/CN, language=CN, createTime=1751182263881, createBy=18614031015, updateTime=1751778001962, updateBy=18614031015, name=科技导报, tplId=1146099689490845704, title=科技导报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148021146403992296, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=articleTextType, value=kx, createTime=1751639170504, updateTime=1751639170504, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146378826469, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=banner, value=null, createTime=1751639170498, updateTime=1751639170498, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146366243556, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=9GHSf7eGlIPH0Tv/OOdstA==, createTime=1751639170495, updateTime=1751639170495, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146395603687, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751639170502, updateTime=1751639170502, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146387215078, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751639170500, updateTime=1751639170500, creator=18614031015, updator=18614031015)]), Website(id=1146105254833139715, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031591421210625, 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/kjdb/EN, language=EN, createTime=1751182386363, createBy=18614031015, updateTime=1753500121937, updateBy=18614031015, name=科技导报, tplId=1146101810881728533, title=Science & Technology Review, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155838567709528217, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=articleTextType, value=kx, createTime=1753502988984, updateTime=1753502988984, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567692750998, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=banner, value=null, createTime=1753502988980, updateTime=1753502988980, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567688556693, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic?fileId=9GHSf7eGlIPH0Tv/OOdstA==, createTime=1753502988979, updateTime=1753502988979, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567705333912, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic, createTime=1753502988983, updateTime=1753502988983, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567701139607, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1753502988982, updateTime=1753502988982, creator=18614031015, updator=18614031015)])], journalTitle=科技导报, weixinUrl=null, journalUrl=null, iacademicId=null, status=1, seqNo=null, journalTitleEn=Science & Technology Review, journalPhotoCn=wfghvu3bhh/dKxuZ+ucVHA==, journalPhotoEn=yjSfclmpNm7ihn9NbTZ69g==, journalFirstLetter=K, journalRecommend=null, journalNew=null, journalCollection=1, jcrJf=null, cjcrJf=0.91, 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/kjdb/CN/10.3981/j.issn.1000-7857.2025.12.00116, detailUrlEn=https://castjournals.cast.org.cn/joweb/kjdb/EN/10.3981/j.issn.1000-7857.2025.12.00116, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kjdb/CN/PDF/10.3981/j.issn.1000-7857.2025.12.00116, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kjdb/EN/PDF/10.3981/j.issn.1000-7857.2025.12.00116, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1774627200000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于滤波与多尺度时序协同的含噪多变量预测
收藏切换
PDF下载
郭启文 1 , 张玉龙 1, * , 张欣 2 , 李德杰 3 , 汪昕 1 , 张海霞 1 , 王宇涵 1
科技导报 | 特色专题 2026,44(6): 48-56
收起
收藏切换
科技导报 |特色专题 2026 , 44 (6) : 48 -56
基于滤波与多尺度时序协同的含噪多变量预测
全屏
[Author(id=1251567849513365577, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2410425@stu.neu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1251567849572085835, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849513365577, language=EN, stringName=Qiwen GUO, firstName=Qiwen, middleName=null, lastName=GUO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567849639194700, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849513365577, language=CN, stringName=郭启文, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio={"content":"

郭启文,博士研究生,研究方向为露天矿粉尘防控,电子信箱:

"}, bioImg=null, bioContent=

郭启文,博士研究生,研究方向为露天矿粉尘防控,电子信箱:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567849693720654, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhangyulong@mail.neu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1251567849752440912, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849693720654, language=EN, stringName=Yulong ZHANG, firstName=Yulong, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567849836326993, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849693720654, language=CN, stringName=张玉龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567849890852947, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, 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=1251567849966350421, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849890852947, language=EN, stringName=Xin ZHANG, firstName=Xin, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2College of Civil Engineering, Shandong Jianzhu University, Jinan 250101, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850025070678, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567849890852947, language=CN, stringName=张欣, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2山东建筑大学土木工程学院,济南 250101, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567849190404162, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=2, ext=[AuthorCompanyExt(id=1251567849202987075, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849190404162, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2College of Civil Engineering, Shandong Jianzhu University, Jinan 250101, China), AuthorCompanyExt(id=1251567849211375684, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849190404162, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2山东建筑大学土木工程学院,济南 250101)])]), Author(id=1251567850079596632, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=3, 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=1251567850150899802, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850079596632, language=EN, stringName=Dejie LI, firstName=Dejie, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3CCCC Second Harbour Engineering Co., Ltd., Wuhan 430040, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850209620059, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850079596632, language=CN, stringName=李德杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3中交第二航务工程局有限公司,武汉 430040, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567849282678853, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=3, ext=[AuthorCompanyExt(id=1251567849320427590, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849282678853, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3CCCC Second Harbour Engineering Co., Ltd., Wuhan 430040, China), AuthorCompanyExt(id=1251567849324621895, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567849282678853, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3中交第二航务工程局有限公司,武汉 430040)])]), Author(id=1251567850268340317, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, 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=1251567850335449183, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850268340317, language=EN, stringName=Xin WANG, firstName=Xin, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850406752352, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850268340317, language=CN, stringName=汪昕, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567850524192866, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=5, 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=1251567850603884644, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850524192866, language=EN, stringName=Haixia ZHANG, firstName=Haixia, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850666799205, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850524192866, language=CN, stringName=张海霞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])]), Author(id=1251567850721325159, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, orderNo=6, 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=1251567850784239721, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850721325159, language=EN, stringName=Yuhan WANG, firstName=Yuhan, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251567850876514410, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, authorId=1251567850721325159, language=CN, stringName=王宇涵, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251567847638511679, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, xref=1, ext=[AuthorCompanyExt(id=1251567847642705984, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China), AuthorCompanyExt(id=1251567847646900289, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1251559135813255893, companyId=1251567847638511679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819)])])]
郭启文1 , 张玉龙1, * , 张欣2, 李德杰3, 汪昕1, 张海霞1, 王宇涵1
作者信息
  • 1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819
  • 2山东建筑大学土木工程学院,济南 250101
  • 3中交第二航务工程局有限公司,武汉 430040
通讯作者:
张玉龙(通信作者),副教授,研究方向为多场耦合岩体力学,电子信箱:
Noisy multivariate prediction via filtering and multi−scale temporal synergy
Qiwen GUO1 , Yulong ZHANG1, * , Xin ZHANG2, Dejie LI3, Xin WANG1, Haixia ZHANG1, Yuhan WANG1
Affiliations
  • 1State Key Laboratory of Intelligent Deep Metal Mining and Equipment, Northeastern University, Shenyang 110819, China
  • 2College of Civil Engineering, Shandong Jianzhu University, Jinan 250101, China
  • 3CCCC Second Harbour Engineering Co., Ltd., Wuhan 430040, China
出版时间: 2026-03-28 doi: 10.3981/j.issn.1000-7857.2025.12.00116
文章导航
收藏切换

含噪多变量预测是环境、交通、工业等领域的共性难题,其核心挑战在于平衡噪声过滤与多尺度特征捕捉。提出一种基于卡尔曼滤波−长短期记忆网络−Transformer的混合模型(Kalman−LSTM−Transformer)。该模型在过滤噪声的同时捕捉局部时序与全局依赖,并结合贝叶斯优化以实现高效精准预测。以露天矿粉尘浓度预测为案例验证,基于1年监测数据的实验显示,该模型较基准模型的均方根误差降低21.70%~27.19%,平均绝对误差降低6.68%~18.30%,决定系数R2达0.934;消融实验与超参数分析结果进一步证实了各模块有效性。该模型可迁移至同类场景,为多领域智能预警与精准治理提供支撑。

含噪多变量预测  /  卡尔曼滤波  /  长短期记忆网络  /  Transformer  /  贝叶斯优化

Noisy multivariate prediction is a common challenge in fields such as environmental science, transportation, and industry. The core difficulty lies in balancing noise filtering with multi−scale feature capture. To address this, a hybrid model (Kalman−LSTM−Transformer) based on Kalman filter, long short−term memory (LSTM), and Transformer is proposed. This model captures local temporal and global dependencies while filtering noise, and integrates Bayesian optimization to achieve efficient and accurate prediction. Using open−pit mine dust concentration prediction as a case study, experiments based on 1−year of monitoring data demonstrate that the model outperforms baseline models, reducing the root mean square error (RMSE) by 21.70%–27.19% and the mean absolute error (MAE) by 6.68%–18.30%, while achieving a coefficient of determination (R2) of 0.934. Ablation experiments and hyperparameter analysis results further confirm the effectiveness of each module. The model exhibits transferability to similar scenarios, providing support for intelligent early warning and precision management across multiple domains.

noisy multivariate prediction  /  Kalman filter  /  long short−term memory  /  Transformer  /  Bayesian optimization
郭启文, 张玉龙, 张欣, 李德杰, 汪昕, 张海霞, 王宇涵. 基于滤波与多尺度时序协同的含噪多变量预测. 科技导报, 2026 , 44 (6) : 48 -56 . DOI: 10.3981/j.issn.1000-7857.2025.12.00116
Qiwen GUO, Yulong ZHANG, Xin ZHANG, Dejie LI, Xin WANG, Haixia ZHANG, Yuhan WANG. Noisy multivariate prediction via filtering and multi−scale temporal synergy[J]. Science & Technology Review, 2026 , 44 (6) : 48 -56 . DOI: 10.3981/j.issn.1000-7857.2025.12.00116
多变量长时序数据的精准预测是环境监测、交通调度、工业预警等领域实现主动防控的核心需求,此类数据普遍面临强噪声干扰、非线性交互及长短期特征耦合的共性挑战[13]。露天矿开采作为全球资源供应的核心支柱,在推动工业发展与经济增长的同时,其粉尘污染因广泛性、持久性和高危害性,已成为威胁人体健康、破坏生态环境及损害生产设备的突出问题[47]。受现场复杂作业环境与多要素交互影响,粉尘浓度时序数据同样存在上述共性难题,导致传统被动治理措施效果有限[89]。由此可见,发展兼顾噪声处理与多尺度特征建模的预测方法,不仅对露天矿粉尘污染防控至关重要,更可为跨领域同类问题提供借鉴。
露天矿粉尘浓度预测研究关键难点在于监测数据本身具有的高度复杂性特征。由粉尘浓度与气象参数共同组成的多变量时序数据序列具有强非线性与非平稳特性,并且表现出显著多尺度变化规律。具体而言,风速与风向直接决定了粉尘的水平迁移路径与扩散通量,是导致污染范围扩大的动力来源;相对湿度通过控制颗粒物的吸湿增长效应使粉尘粒径与自重增大,从而加速干湿沉降过程;而气温与气压的变化与大气稳定度和逆温层的形成紧密关联,从而影响粉尘在垂直方向上的扩散能力。这种复杂的多物理场交互作用意味着单一维度的历史浓度数据难以完整表征系统演化状态,必须融合多维气象信息以重构粉尘扩散的动力学过程。更为复杂的是,这些时序数据还伴有开采现场复杂环境条件引起的严重噪声污染影响[10]。在实际数据采集过程中,传感器误差、设备机械振动、信号传输失真等会引入观测噪声;同时,气象参数与生产活动的突发性变化会导致粉尘浓度产生高频、非平稳的波动,使得时序数据呈现复杂的“类噪声”特征,导致真实的浓度变化规律变得模糊不清[11],特别是对于突发峰值信号的表征能力产生了明显的削弱作用。传统的移动平均处理方法与小波分析技术等预处理手段往往难以在滤除观测噪声的同时,有效保留并解析表征真实物理过程的突发波动,导致峰值预测结果出现时间延迟或形态失真的情况频繁发生[12]。这种噪声与有效特征相互耦合的复杂效应成为制约预测模型精度提升的关键瓶颈,开发具有自适应特性的新型噪声过滤技术显得尤为必要,这将为后续建模工作提供更为可靠的数据基础。
近年来,由数据驱动主导的机器学习方法在粉尘预测领域取得显著进展性成果,如基于支持向量机[13]、随机森林[14]及其集成模型[15]的模型框架在特定场景粉尘预测任务中均表现出优于传统模型的预测性能。但对于露天矿多变量长时序粉尘预测任务而言,这些方法难以实现有效捕捉特征非线性交互关系以及建立粉尘扩散长期依赖规律的需求[1617]
深度学习技术的快速发展为解决上述问题提供了新的可能性。通过门控机制的设计,长短期记忆网络(long short−term memory,LSTM)及其衍生架构能够较好地捕捉时序短期依赖关系,这一特点使其在粉尘浓度预测研究中获得初步应用成功[1819]。实例表明,露天矿粉尘浓度受气象要素长期累积效应的影响显著,然而LSTM类模型在处理长程依赖时面临记忆衰减现象的困扰[20]。具有自注意力机制的Transformer模型理论上具备更强的长程建模潜力[21],但在实际粉尘预测应用中存在2方面制约因素:标准注意力机制对局部突变特征的敏感性不足造成峰值预测响应延迟;二次方计算复杂度问题阻碍了其在长序列数据中的高效应用[22]。现有混合模型虽尝试整合空间与时序特征,但普遍存在数据噪声负面影响被忽视的问题,同时伴随较高的模型结构复杂度,进而影响训练效率。
为克服上述研究不足,提出一种基于卡尔曼滤波增强的LSTM−Transformer混合模型(Kalman−LSTM−Transformer),从3个方面实现预测精度与效率的同步提高:首先在卡尔曼滤波器中引入动态噪声协方差调整机制,满足滤除多源噪声的同时保留突发性峰值特征,最终实现自适应噪声过滤;其次,在利用LSTM捕捉局部时序模式的基础上,通过Transformer编码器建模全局长期依赖;最后,采用贝叶斯优化算法,控制模型计算成本的同时提升其泛化能力。
Kalman−LSTM−Transformer的整体架构以“去噪−局部建模−全局建模”为核心逻辑,主要包括卡尔曼滤波、LSTM特征提取与Transformer全局建模3个核心模块。该架构设计针对含噪声多变量长时序预测的共性需求,核心模块具有场景无关性。原始监测数据首先经卡尔曼滤波处理来抑制传感器噪声并削弱瞬时异常,降噪后的数据随后被重构为时间序列并输入LSTM,以捕捉短期的时序依赖。LSTM输出的局部特征通过投影与位置编码后,输入至Transformer编码器完成进一步长期依赖关系建模,最后利用全连接层输出粉尘浓度预测值。图1展示了该模型整体预测流程,3大核心模块的协同作用关系可表示为
$ \widehat {{y_t}} = {{\text{F}}_\theta }({\text{Transformer}}({\text{LSTM}}({\text{Kalman}}({{X}_{{t - se}{{{q}}_{{{len}}}}{{:t - 1}}}})))) $
式中,$ {{{X}}_{{{t - se}}{{{q}}_{{{len}}}}{{:t - 1}}}} $表示历史输入序列,$ {\text{Kalman}}( \cdot ) $$ {\text{LSTM}}( \cdot ) $$ {\text{Transformer}}( \cdot ) $分别表示3个核心模块的映射函数,$ {{\text{F}}_\theta } $为带参数$ \theta $的全连接层,$ \widehat {{y_t}} $t时刻的粉尘浓度预测值。在模型预测流程中,卡尔曼滤波器主要作为独立的数据预处理模块以接受含噪声的原始数据,其输出结果为噪声净化后的时序数据。该时序数据后续将作为输入序列进入由LSTM和Transformer组成的特征学习网络进行预测,确保特征学习阶段建立在洁净数据基础之上,从而提升整体预测性能。
原始粉尘浓度数据常受传感器噪声与环境干扰影响,具有非平稳性与突发性波动特征,需通过预处理保障后续模型的建模精度。卡尔曼滤波作为一种最优递归数据处理算法,通过对系统状态的递推估计,可有效融合含噪声观测数据与系统动态模型,实现对真实信号的最优估计。其在本模型中的作用机制为定义状态方程和观测方程
$ \left\{ \begin{gathered} {{{x}}_{k}} = {{A}}{{{x}}_{k - 1}}+{{B}}{{{u}}_{k - 1}}+{{{w}}_{k - 1}} \hfill \\ {{{z}}_k} = {{H}}{{{x}}_{k}}+{{{v}}_{k}} \hfill \\ \end{gathered} \right. $
式中,$ {{{x}}_{k}} $表示k时刻的状态向量,$ {{{z}}_{k}} $表示k时刻的观测向量,A为状态转移矩阵,B为控制矩阵,H为观测矩阵,$ {{{w}}_{k}}\sim {{N(0,Q)}} $$ {{{v}}_{k}}\sim {{N(0,R)}} $分别代表过程噪声和观测噪声(QR为协方差矩阵)。为克服固定噪声协方差对粉尘浓度突发波动跟踪能力不足的问题,参考指数平滑思想,该模型利用自适应更新机制动态调整R的取值范围:
$ {R_k} = \alpha \cdot {R_{k - 1}}+(1 - \alpha ) \cdot e_k^2 $
式中,$ {e_k} $表示预测误差,$ \alpha $为平滑系数,用于平衡历史估计与当前信息的权重。本研究基于数据波动特性的先验分析,设定$ \alpha = 0.9 $。该取值旨在使观测噪声协方差$ {R_k} $既能保持稳定,又可对突发性预测误差做出灵敏调整,从而增强模型对真实浓度突变的跟踪能力。这种设计保证了滤波器在面对持续的观测噪声和粉尘浓度真实突发波动时实现滤波强度自适应调整,从而在抑制噪声与保留有效突变特征之间取得更优平衡。
经卡尔曼滤波处理后的粉尘浓度序列被输入LSTM网络以捕捉局部时序依赖,其结构如图2所示。LSTM通过门控机制选择性保留或遗忘时序信息,有效捕捉非平稳序列的短期动态特征,其完整处理过程可表示为[23]
$ \begin{gathered} {f_t} = \sigma ({{{W}}_{f}} \cdot [{h_{t - 1}},{x_t}]+{{{b}}_{f}}) \hfill \\ {i_t} = \sigma ({{{W}}_i} \cdot [{h_{t - 1}},{x_t}]+{{{b}}_{i}}) \hfill \\ \widetilde {{c_t}} = \tanh ({{{W}}_{c}} \cdot [{h_{t - 1}},{x_t}]+{{{b}}_{c}}) \hfill \\ {c_t} = {f_t} \odot {c_{t - 1}}+{i_t} \odot \widetilde {{c_t}} \hfill \\ {o_t} = \sigma ({{{W}}_{o}} \cdot [{h_{t - 1}},{x_t}]+{{{b}}_{o}}) \hfill \\ {h_t} = {o_t} \odot \tanh \left( {{c_t}} \right) \hfill \\ \end{gathered} $
式中,$ {f_t} $$ {i_t} $$ {o_t} $分别表示遗忘门、输入门和输出门;$ \widetilde {{c_t}} $为候选细胞状态,$ {c_t} $表示更新后细胞状态;$ {h_t} $为输出的隐藏状态;$ {{{W}}_{f}} $$ {{{W}}_{i}} $$ {{{W}}_{c}} $$ {{{W}}_{o}} $分别为遗忘门、输入门、候选细胞状态和输出门权重矩阵,$ {{{b}}_{f}} $$ {{{b}}_{i}} $$ {{{b}}_{c}} $$ {{{b}}_{o}} $为对应上述权重矩阵的偏置项。
降噪后时序数据输入LSTM后,遗忘门首先会依据当前输入与上一时刻隐藏状态来选出需要保留的历史细胞状态信息,并且剔除冗余信息的干扰;输入门用于判定当前时刻的新特征是否需要加入细胞状态;候选细胞状态利用非线性变换完成当前时刻的潜在状态特征的生成,将其与上述保留的历史信息融合后形成更新后细胞状态;最后输出门结合细胞状态与当前输入生成包含局部时序特征的隐藏状态并传递至下一时刻。上述完整的动态筛选机制使LSTM能够精准捕捉相邻时刻的浓度波动关联以及瞬时动态变化,为后续Transformer全局建模提供细粒度的局部特征支撑。为实现模型复杂度与特征提取能力的动态平衡,本研究构建了双层堆叠的LSTM结构。第1层用于将高维输入序列映射为初步的时序特征表示,第2层则进一步提取深层的非线性模式。每层的隐藏单元数统一设定为64,保证模型具备足够的记忆特性和抗过拟合风险能力,确保了局部特征提取的鲁棒性。
LSTM输出的局部特征经线性投影转换为统一维度后,引入位置编码来保留有效时序信息。随后输入至Transformer编码器完成全局依赖捕捉,其详细组成架构如图3所示。Transformer通过多头自注意力机制建模长期依赖,其核心是为序列中不同时间步的特征分配注意力权重,计算公式为
$ Attention({{Q}},{{K}},{{V}}) = softmax(\frac{{{{Q}}{{{K}}^{\rm{T}}}}}{{\sqrt {{{{d}}_{{k}}}} }}){{V}} $
式中,QKV分别表示查询、键、值矩阵,$ {{{d}}_{k}} $表示特征维度。多头注意力通过并行计算多个注意力头并拼接结果,增强特征表达能力[24]
$ MultiHead({{Q}},{{K}},{{V}}) = Concat(hea{d_1},...,hea{d_h}){{{W}}^{O}} $
式中,$ head_i={Attention}({QW}_i^Q,{KW}_i^K,{VW}_i^V) $$ {{W}}_{i}^{Q} $$ {{W}}_{i}^{K} $$ {{W}}_{i}^{V} $表示投影矩阵,$ {{{W}}^{O}} $为输出矩阵。
多头自注意力机制根据不同时间步特征相似度计算结果,对其中的关键时间步分配更高权重,实现长期依赖捕捉。为了保证模型能够从不同子空间并行捕捉序列的全局关联信息,本研究将多头注意力机制的头数设定为4,同时为了增强模型的非线性映射能力,前馈网络层的中间维度设定为256。前馈网络则利用非线性变换进一步实现特征提取,并通过残差连接来避免训练过程中出现梯度消失,层归一化用于稳定训练过程中的特征分布。上述处理方式有效补充了局部特征的全局关联信息,实现粉尘浓度短期波动与长期趋势的双重捕捉,从而提高预测精度。
为验证上述模型在含噪声多变量长时序预测任务中的有效性,以露天矿粉尘浓度预测为具体案例,采用内蒙古某高寒高海拔金属露天矿区的监测数据。为全面评估现场露天环境下粉尘浓度的动态变化,采取现场连续监测的方式开展了2024年全年的时序数据监测,监测频率1 h,包括了粉尘浓度和露点温度、温度等7项气象参数。受设备故障或环境干扰影响,原始数据存在局部缺失,经均值填充预处理得到8736组有效数据,基本统计量如表1所示。
在该多变量长时序粉尘浓度预测任务中,利用过去12 h的监测数据预测未来1 h的粉尘浓度。针对风向这一具备周期性的离散特征,为使其方向信息能被模型有效理解并避免引入错误的序数关系,本研究采用正弦−余弦编码方法将其转换为2个连续的、具有周期性的特征维度,并与其他连续气象参数统一开展归一化处理。为了消除特征量纲差异对模型训练的干扰,采用最大−最小标准化方法对8类特征进行归一化处理,将数据压缩至(0,1)区间内,从而提升训练稳定性。所有特征序列按照7∶1∶2的比例划分为训练集、验证集与测试集。该时序划分策略既保留了数据的时间依赖性,可有效捕捉粉尘浓度与气象因子的滞后关联,又能规避随机划分导致的时序信息泄露,保障评估结果的可靠性。
为系统验证本研究提出模型在去噪增强与多尺度特征融合的创新性,选取3类代表性时序预测基准模型开展对比实验:(1) LSTM作为经典循环神经网络(recurrent neural network,RNN)的变体,其门控机制可有效建模多变量时间序列的短期依赖关系,且对输入特征的尺度差异具有鲁棒性;(2) Seq2Seq通过编码器−解码器框架完成序列到序列的映射,能够灵活处理多变量输入与输出之间的非线性关系;(3) Transformer利用自注意力机制捕捉多变量之间的全局依赖关系,突破了传统RNN模型的串行计算瓶颈,尤其擅长处理长序列数据。
为减少人工干预并降低时间成本,本研究采用贝叶斯优化算法开展超参数寻优,在预设超参数空间中自适应搜索最优参数配置。具体的寻优对象主要包括对模型性能影响显著的关键超参数,如学习率(0.0001~0.01)、批次大小(16~128)、LSTM隐藏层节点数(32~128)以及Dropout率(0.1~0.5)。优化算法以平衡模型性能与计算开销为前提,通过高斯过程回归模型不断修正后验证概率分布,最终确定最优超参数组合。训练过程中,选取均方误差(mean−square error,MSE)作为损失函数,通过衡量预测值与真实值的均方差指导模型参数更新,驱动网络学习粉尘浓度序列的潜在模式。需要特别说明的是,由于输入数据经过了归一化处理,直接计算误差无法反映真实场景下的物理偏差。因此,在模型输出预测结果后,利用最大−最小标准化的逆变换公式,严格对预测值与真实标签进行反归一化操作,将其恢复至原始粉尘浓度量纲(μg/m3),确保评价指标具有明确的工程物理意义,能够真实反映模型在实际矿山环境中的预测精度。为全面评估模型预测效果,采用平均绝对误差(mean absolute error,MAE)、均方根误差(root mean square error,RMSE)和决定系数(R2)作为评价指标。损失函数与评价指标的计算公式为
$ F_{\text{MSE}}=\frac{1}{N}\sum\limits_{t=1}^N\left(y_\mathrm{o}-y\mathrm{_p}\right)^2 $
$ F_{\text{MAE}}=\frac{1}{N}\sum\limits_{t=1}^N\left|y_\mathrm{o}-y_\mathrm{p}\right| $
$ F_{\text{RMSE}}=\sqrt{\frac{1}{N}\sum\limits_{t=1}^N\left(y\mathrm{_o}-y\mathrm{_p}\right)^2} $
$ {F_{{{\text{R}}^{\text{2}}}}} = 1 - \frac{{\displaystyle\sum\limits_{t = 1}^N {{{\left( {{y_{\mathrm{o}}} - {y_{\mathrm{p}}}} \right)}^2}} }}{{\displaystyle\sum\limits_{t = 1}^N {{{\left( {{y_{\mathrm{o}}} - {{\bar y}_{\mathrm{p}}}} \right)}^2}} }} $
式中,yo表示粉尘浓度预测值,yp表示粉尘浓度监测值,$\bar y_{\mathrm{p}} $表示粉尘浓度监测值的均值,N表示总样本数量。
本研究首先采用Kalman−LSTM−Transformer模型开展多变量长时序粉尘浓度预测,与参数基准一致的基准模型进行对比实验,以评价其整体预测性能的优异性;其次通过消融实验验证去噪增强模块与多尺度特征融合模块在多变量长时序粉尘浓度预测中的有效性;最后对比分析不同超参数优化方式下模型的精度与计算效率。
图4显示了不同模型预测结果与真实值对比情况,并展示了对应模型相对误差的分布特征。由图4(a)可知,各模型在粉尘浓度长期趋势预测中均表现出较稳定的性能,即在全局监测区间内可精准捕捉浓度整体波动特征。但针对局部放大图中高频波动区间,Kalman−LSTM−Transformer模型与3种基准模型的性能表现出较大差异。LSTM、Seq2Seq及Transformer模型在高频峰值区域的预测值与真实值偏差较大,表明其在捕获高频区域复杂时间依赖性及突发变化上存在局限性。Kalman−LSTM−Transformer模型在该区域则表现优异,预测值与真实值吻合度高,证明信号降噪与多尺度特征融合发挥了关键作用。
为定量直观地展示所提模型的优越性,图4(b)对比了不同模型预测结果的相对误差。4种模型的相对误差均呈正态分布,但分布范围差异显著。Kalman−LSTM−Transformer模型的相对误差集中于0附近且密度最高,而其他3种模型的误差分布范围明显更大。表2列出了各模型的评价指标结果,3种基准模型的RMSE为38.821~41.751,MAE为21.851~24.959,R2为0.874~0.892;相比之下,Kalman−LSTM−Transformer模型的RMSE降至30.398(降低21.70%~27.19%),MAE降至20.391(降低6.68%~18.30%),R2提升至0.934(提高4.71%~6.86%)。模型性能的显著提升归因于Kalman−LSTM−Transformer各模块的协同设计与优势互补。RMSE显著降低主要得益于卡尔曼滤波模块有效过滤了非平稳序列中的高频随机噪声,降低了预测误差的方差;而R2提升则验证了Transformer模块在长序列中成功捕获了气象因子与粉尘浓度间的全局依赖关系,修正了单一LSTM模型在峰值处的滞后现象。这种集精准去噪、局部精细建模和全局关系捕捉于一体的联合架构,让模型能够同时应对噪声干扰、短期突变和长期趋势,从而在复杂的多变量长时序预测任务中展现出更全面、稳健且精准的预测能力。综上,Kalman−LSTM−Transformer模型的综合预测性能优于3种典型时序基准模型,具有更高的稳定性与精度,更适用于多变量长时序粉尘浓度预测任务。
为有效说明Kalman−LSTM−Transformer模型中各核心模块的重要作用,分别设计3组对照模型以开展消融实验,用于量化各模块对预测性能的贡献程度。
1) LSTM−Transformer。移除卡尔曼滤波模块,直接将原始数据输入LSTM与Transformer,用于验证卡尔曼滤波在噪声抑制与数据质量优化中的作用,以及对模型抗干扰能力的影响。
2) Kalman−LSTM。移除Transformer模块,仅保留卡尔曼滤波与LSTM,旨在评估Transformer捕捉长时序全局依赖的必要性,分析其对跨时间步周期性特征建模的贡献。
3) Kalman−Transformer。移除LSTM模块,保留卡尔曼滤波与Transformer,以验证LSTM在捕捉局部短期波动中的关键作用,观察其对瞬时动态预测精度的影响。
图5展示了Kalman−LSTM−Transformer模型及其消融模型的粉尘浓度预测相对误差变化趋势与范围。由图5(a)可见,LSTM−Transformer的误差波动剧烈;Kalman−LSTM借助卡尔曼滤波抑制噪声,对局部短时波动的拟合更稳定,但长程依赖捕捉能力不足。深入分析局部峰值区域时间节点可知,引入单一机制的Kalman−LSTM与Kalman−Transformer虽然在一定程度上缩小了误差的震荡边界,但在面对产生剧烈波动的突发时刻时,仍表现出较大幅度的偏差,说明单一的局部或全局建模难以完全捕捉复杂的动态演化规律。相比之下,Kalman−LSTM−Transformer集成模型的误差曲线最为平稳,在整个监测周期内大多围绕零轴波动,即便在突变时刻也能保持较低的误差水平,兼顾了短时灵敏性与长程稳定性,证明了该模型通过协同降噪与多尺度特征提取,有效增强了系统的抗干扰能力与鲁棒性,能够在复杂多变的工况下保持高度稳定的预测精度。图5(b)进一步揭示Kalman−LSTM与Kalman−Transformer的误差箱体宽度及基于1.5倍四分位距(1.5IQR)的离散程度介于LSTM−Transformer与融合模型之间,表明二者分别强化了局部或全局建模能力,但未实现协同增益。结合表3的评价指标对比结果可知,Kalman−LSTM、Kalman−Transformer的评价指标均优于LSTM−Transformer,却显著差于融合模型。综上可知,卡尔曼滤波降噪为基础支撑,LSTM捕捉短时关联与Transformer建模长程依赖形成互补,三者协同作用使误差分布更集中、预测精度更优,印证了模块融合对复杂多变量长时序预测的关键价值。
本研究选择随机网格搜索、粒子群优化、鲸鱼优化与贝叶斯优化4种方法开展对比实验。图6展示了不同优化策略下模型预测评价指标结果。可以看出随机网格搜索的RMSE与MAE数值最高、R2最低,说明训练过程中随机性导致参数组合未能实现最优解;粒子群优化与鲸鱼优化的评价指标优于随机网格搜索,但相比于贝叶斯优化仍有不足。贝叶斯优化策略基于高斯过程的智能搜索实现对参数关联的精准捕捉,从而实现更高的预测精度。此外,针对优化效率而言,与随机网格搜索、粒子群优化等算法相比,贝叶斯优化基于高斯过程构建代理模型以引导搜索方向,能够以更少的模型评估次数收敛到性能优异的参数区域。这体现了其在平衡未知区域与利用已知优势的智能性,保证寻优效果的同时有效降低了超参数调优的时间与计算成本。综上所述,贝叶斯优化策略能更高效地挖掘超参数空间潜力,从而显著提升模型的最终预测精度与稳定性。
本研究针对复杂环境下含噪多变量长时序预测面临的共性挑战,提出了一种集成自适应卡尔曼滤波、LSTM与Transformer的混合预测模型。通过机制融合有效解决噪声过滤与关键特征保留难以兼顾的难题,即依托卡尔曼滤波动态抑制观测噪声,凭借LSTM与Transformer精准捕捉局部波动与全局依赖,并结合贝叶斯优化实现参数适配。基于内蒙古某露天矿全年实测数据的验证结果表明,该模型在各项评价指标上均优于LSTM、Seq2Seq及Transformer等主流基准模型,其中RMSE降低21.70%~27.19%,MAE降低了6.68%~18.30%,R2提升了4.71%~6.86%。这一结果不仅说明了模型在矿山粉尘预警中的有效性,更证实了其能够在不损失关键信息的条件下有效处理高噪声数据,降低实际工程应用中对数据质量的高要求。
虽然本研究聚焦于露天矿粉尘预测场景展开验证,但该模型架构在处理含噪声、非线性及长短期特征耦合的时序数据方面展现出了一定的通用性潜力,为环境、交通及工业等领域提供了一种低成本、高精度的数据驱动治理范式。针对特定领域的数据特性,完成输入特征与模型参数适配调整后,即有望推广至环境监测、交通流量预测等同类任务中。未来的研究工作将重点关注动态噪声协方差机制的进一步优化,并探索构建多站点协同的时空耦合预测模型,期望提升模型在复杂开放环境下的鲁棒性,从而推动智能监测系统从单一节点预测向区域协同决策转变。
  • 国家重点研发计划项目(2023YFC2907303)
  • 国家自然科学基金面上项目(12372374)
参考文献 引证文献
排序方式:
[1]
毛玉如, 汪赟, 张建勋. 中国噪声自动监测的现状和发展[J]. 科技导报, 2024, 42(20): 23-31.
[2]
袁旻忞, 王彦琴, 邵社刚, . 道路交通噪声控制技术研发进展[J]. 科技导报, 2024, 42(20): 85-94.
[3]
户文成, 刘磊, 康钟绪, . 城市轨道交通噪声污染防治进展[J]. 科技导报, 2024, 42(20): 70-84.
[4]
Entwistle J A, Hursthouse A S, Marinho Reis P A, et al. Metalliferous mine dust: Human health impacts and the potential determinants of disease in mining communities[J]. Current Pollution Reports, 2019, 5(3): 67-83.
[5]
Hosseini S, Poormirzaee R, Gilani S O, et al. A reliability−based rock engineering system for clean blasting: Risk analysis and dust emissions forecasting[J]. Clean Technologies and Environmental Policy, 2023, 25(6): 1903-1920.
[6]
Hendryx M, Islam M S, Dong G H, et al. Air pollution emissions 2008−2018 from Australian coal mining: Implications for public and occupational health[J]. International Journal of Environmental Research and Public Health, 2020, 17(5): 1570.
[7]
Guo X D, Sun Q F, Zhao Y S. et al. Identification and characterisation of heavy metals in farmland soil of Hunchun basin[J]. Environmental Earth Sciences, 2019, 78(10): 310.
[8]
Zhang H, Han W H, Xu Y L, et al. Analysis on the development status of coal mine dust disaster prevention technology in China[J]. Journal of Healthcare Engineering, 2021: 5574579.
[9]
Li X, Zhao H, Cheng J Y, et al. BWO−BiLSTM & CNN composite model for prediction of atmospheric particulate matter mass concentration[J]. Atmospheric Pollution Research, 2024, 15(11): 102273.
[10]
Chen Y C, Li D C. Selection of key features for PM2.5 prediction using a wavelet model and RBF−LSTM[J]. Applied Intelligence, 2021, 51(4): 2534-2555.
[11]
Yadav V, Yadav A K, Singh V, et al. Artificial neural network an innovative approach in air pollutant prediction for environmental applications: A review[J]. Results in Engineering, 2024, 22: 102305.
[12]
温雯, 胡智鑫, 郝志峰. 面向序列推荐的深度化指数移动平均学习方法[J]. 计算机科学与探索, 2025, 19(3): 774-786.
[13]
Leong W C, Kelani R O, Ahmad Z. Prediction of air pollution index (API) using support vector machine (SVM)[J]. Journal of Environmental Chemical Engineering, 2020, 8(3): 103208.
[14]
Kamran Kamran, Jiskani I M, Wang Z, et al. Decision intelligence−driven predictive modelling of air quality index in surface mining[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 108399.
[15]
Yang Y K, Zhou W, Wang Z M, et al. Accurate long−term dust concentration prediction in open−pit mines: A novel machine learning approach integrating meteorological conditions and mine production intensity[J]. Journal of Cleaner Production, 2024, 436: 140411.
[16]
Di Q, Amini H, Shi L H, et al. An ensemble−based model of PM2.5 concentration across the contiguous United States with high spatiotemporal resolution[J]. Environment International, 2019, 130: 104909.
[17]
Choubin B, Abdolshahnejad M, Moradi E, et al. Spatial hazard assessment of the PM10 using machine learning models in Barcelona, Spain[J]. Science of The Total Environment, 2020, 701: 134474.
[18]
Li P, Wang S W, Ji H, et al. Air quality index prediction based on an adaptive dynamic particle swarm optimized bidirectional gated recurrent neural network–China region[J]. Advanced Theory and Simulations, 2021, 4(12): 2100220.
[19]
Liu B C, Yu Z C, Wang Q S, et al. Prediction of SSE Shanghai Enterprises index based on bidirectional LSTM model of air pollutants[J]. Expert Systems with Applications, 2022, 204: 117600.
[20]
Ding C, Zheng Z Y, Zheng S R, et al. Accurate air−quality prediction using genetic−optimized gated−recurrent−unit architecture[J]. Information, 2022, 13(5): 223.
[21]
Zhong J D, Liu J L. Seq2seq modelling for cross−site temporal forecasting of urban air pollutant concentrations leveraging sensor data[J]. Building and Environment, 2025, 269: 112463.
[22]
Chen Y B, Chen X M, Xu A L, et al. A hybrid CNN−Transformer model for ozone concentration prediction[J]. Air Quality, Atmosphere & Health, 2022, 15(9): 1533-1546.
[23]
Gong Y, Wang Y Z, Xie Y H, et al. Dynamic fusion LSTM−Transformer for prediction in energy harvesting from human motions[J]. Energy, 2025, 327: 136192.
[24]
Chen D Y, Liu H. Integrating multi−dimensional graph attention networks and transformer architecture for predicting air pollution in subway stations[J]. Applied Soft Computing, 2025, 178: 113033.
2026年第44卷第6期
PDF下载
460
213
引用本文
BibTeX
文章信息
doi: 10.3981/j.issn.1000-7857.2025.12.00116
  • 接收时间:2025-12-10
  • 首发时间:2026-04-16
  • 出版时间:2026-03-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-12-10
  • 修回日期:2026-02-28
基金
国家重点研发计划项目(2023YFC2907303)
国家自然科学基金面上项目(12372374)
作者信息
    1东北大学深部金属矿智能开采与装备全国重点实验室,沈阳 110819
    2山东建筑大学土木工程学院,济南 250101
    3中交第二航务工程局有限公司,武汉 430040

通讯作者:

张玉龙(通信作者),副教授,研究方向为多场耦合岩体力学,电子信箱:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kjdb/CN/10.3981/j.issn.1000-7857.2025.12.00116
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

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

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