Article(id=1149781955945919116, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2403249, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1714579200000, receivedDateStr=2024-05-02, revisedDate=1734883200000, revisedDateStr=2024-12-23, acceptedDate=null, acceptedDateStr=null, onlineDate=1752058980212, onlineDateStr=2025-07-09, pubDate=1743091200000, pubDateStr=2025-03-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752058980212, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752058980212, creator=13701087609, updateTime=1752058980212, updator=13701087609, issue=Issue{id=1149781952959574654, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='9', pageStart='3529', pageEnd='3967', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752058979501, creator=13701087609, updateTime=1776333392421, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1251596220226027613, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1251596220226027614, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=3896, endPage=3904, ext={EN=ArticleExt(id=1149781956122079886, articleId=1149781955945919116, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Vehicle Fuel Consumption Prediction Method Based on Hyperband-CNN-BiLSTM Model, columnId=1156262728772735295, journalTitle=Science Technology and Engineering, columnName=Papers·Traffics and Transportations, runingTitle=null, highlight=null, articleAbstract=

In order to effectively predict the fuel consumption of vehicles, improve fuel economy and promote energy saving and emission reduction, a Hyperband-CNN-BiLSTM-based motor vehicle fuel consumption prediction method was proposed. Firstly, based on the vehicle operating status data and fuel consumption data collected from the actual road test, the salient factors affecting the fuel consumption of vehicles were analyzed. Secondly, combining the powerful feature extraction capability of convolutional neural network(CNN) and the advantages of bidirectional long and short-term memory network (BiLSTM) in dealing with the time-series data, a combined model of vehicle fuel consumption prediction based on CNN-BiLSTM was constructed. Then, in order to improve the model prediction accuracy, the combined model was optimized by Hyperband optimization algorithm, and the vehicle fuel consumption influencing factors were taken as the model input features to train the model to realize the modeling and prediction of vehicle fuel consumption. Finally, CNN, LSTM, BiLSTM, CNN-LSTM and CNN-BILSTM were selected as comparison models to evaluate the effect of Hyperband-CNN-BiLSTM prediction model. The results show that compared with other models, the Hyperband-CNN-BiLSTM model has the smallest mean absolute error (MAE) and root mean squared error (RMSE). They are 0.057 69 and 0.119 25, respectively. R2 is the largest (0.991 76), and the model has the best prediction effect.

, correspAuthors=Mamaiti TURSON, 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, authorCompany=null, fund=null, authors=null, authorsList=Mamaiti TURSON, Hui SUN, Ya-lou LIU), CN=ArticleExt(id=1149781982827213803, articleId=1149781955945919116, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于Hyperband-CNN-BiLSTM模型的车辆油耗预测方法, columnId=1156262730664366426, journalTitle=科学技术与工程, columnName=论文·交通运输, runingTitle=null, highlight=null, articleAbstract=

为了有效地预测车辆的燃油消耗,提高燃油经济性并推动节能减排,提出一种基于Hyperband-CNN-BiLSTM的机动车油耗预测方法。首先基于实际道路测试收集到的车辆运行状态数据和油耗数据,分析了影响车辆油耗的显著性因素;其次结合卷积神经网络(convolutional neural network,CNN)强大的特征提取能力和双向长短期记忆网络(bidirectional long short-term memory,BiLSTM)在处理时序数据方面的优势,构建了基于CNN-BiLSTM的车辆油耗预测组合模型;然后,为提高模型预测准确性,通过Hyperband优化算法对组合模型进行优化,并将车辆油耗影响因素作为模型输入特征,对模型进行训练,实现对车辆油耗的建模和预测;最后,选取CNN、LSTM、BiLSTM、CNN-LSTM、CNN-BiLSTM作为对比模型,对Hyperband-CNN-BiLSTM预测模型效果进行评价。结果表明,相较于其他模型,Hyperband-CNN-BiLSTM模型的平均绝对误差(mean absolute error,MAE)和均方根误差(root mean squared error,RMSE)最小,分别为0.057 69和0.119 25,R2最大,为0.991 76,模型预测效果最佳。

, correspAuthors=吐尔逊·买买提, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=r3xkb4CymV6D3aqYsd2Puw==, magXml=XUYJJBzICWP5PfBUdaAyUA==, pdfUrl=null, pdf=GVASUenw0ARK+7pH7lviEA==, pdfFileSize=5965313, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=fF5r16ba5ikYpn1XbpJydQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=ub29Wy5iUDFhpNd3683lKA==, mapNumber=null, authorCompany=null, fund=null, authors=

吐尔逊·买买提(1978—),男,维吾尔族,新疆阿克苏人,博士,副教授。研究方向:交通环境、数据挖掘。E-mail:

, authorsList=吐尔逊·买买提, 孙慧, 刘亚楼)}, authors=[Author(id=1251595993377092034, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=Tursun@xjau.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1251595993456783812, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, authorId=1251595993377092034, language=EN, stringName=Mamaiti TURSON, firstName=Mamaiti, middleName=null, lastName=TURSON, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251595993553252805, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, authorId=1251595993377092034, language=CN, stringName=吐尔逊·买买提, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio={"content":"

吐尔逊·买买提(1978—),男,维吾尔族,新疆阿克苏人,博士,副教授。研究方向:交通环境、数据挖掘。E-mail:

"}, bioImg=null, bioContent=

吐尔逊·买买提(1978—),男,维吾尔族,新疆阿克苏人,博士,副教授。研究方向:交通环境、数据挖掘。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251595993297400254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, xref=null, ext=[AuthorCompanyExt(id=1251595993305788863, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1251595993314177472, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])]), Author(id=1251595993616167367, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, 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=1251595993695859149, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, authorId=1251595993616167367, language=EN, stringName=Hui SUN, firstName=Hui, middleName=null, lastName=SUN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251595993800716753, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, authorId=1251595993616167367, language=CN, stringName=孙慧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251595993297400254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, xref=null, ext=[AuthorCompanyExt(id=1251595993305788863, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1251595993314177472, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])]), Author(id=1251595993880408534, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, 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=1251595993985266139, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, authorId=1251595993880408534, language=EN, stringName=Ya-lou LIU, firstName=Ya-lou, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251595994043986397, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, authorId=1251595993880408534, language=CN, stringName=刘亚楼, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251595993297400254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, xref=null, ext=[AuthorCompanyExt(id=1251595993305788863, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1251595993314177472, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])])], keywords=[Keyword(id=1251595994190787042, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, orderNo=1, keyword=Hyperband), Keyword(id=1251595994287256037, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, orderNo=2, keyword=fuel rate prediction), Keyword(id=1251595994345976297, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, orderNo=3, keyword=convolutional neural networks (CNN)), Keyword(id=1251595994404696555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, orderNo=4, keyword=bidirectional long short-term memory network (BiLSTM)), Keyword(id=1251595994475999725, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, orderNo=5, keyword=combination model), Keyword(id=1251595994538914289, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, orderNo=1, keyword=Hyperband), Keyword(id=1251595994643771891, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, orderNo=2, keyword=油耗预测), Keyword(id=1251595994719269365, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, orderNo=3, keyword=卷积神经网络(CNN)), Keyword(id=1251595994811544057, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, orderNo=4, keyword=双向长短期记忆网络(BiLSTM)), Keyword(id=1251595994962539003, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, orderNo=5, keyword=组合模型)], refs=[Reference(id=1251595997344903736, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml, language=null, rfNumber=[1], rfOrder=0, authorNames=中华人民共和国生态环境部, journalName=null, refType=null, unstructuredReference=中华人民共和国生态环境部. 中国移动源环境管理年报(2023年)[EB/OL]. (2023-12-07)[2024-05-02]. https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml., articleTitle=中国移动源环境管理年报(2023年), refAbstract=null), Reference(id=1251595997424595514, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml, language=null, rfNumber=[1], rfOrder=1, authorNames=Ministry of Ecology and Environment of the People’s Republic of China, journalName=null, refType=null, unstructuredReference=Ministry of Ecology and Environment of the People’s Republic of China. China mobile source environmental management annual report (2023)[EB/OL]. (2023-12-07)[2024-05-02]. https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml., articleTitle=China mobile source environmental management annual report (2023), refAbstract=null), Reference(id=1251595997495898684, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2017, volume=55, issue=null, pageStart=127, pageEnd=141, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Wang J, Rakha H A, journalName=Transportation Research Part D: Transport and Environment, refType=null, unstructuredReference=Wang J, Rakha H A. Fuel consumption model for heavy duty diesel trucks: model development and testing[J]. Transportation Research Part D: Transport and Environment, 2017, 55: 127-141., articleTitle=Fuel consumption model for heavy duty diesel trucks: model development and testing, refAbstract=null), Reference(id=1251595997554618942, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=Chupp W, Zubrow A, Raymond A, journalName=Washington D C, refType=null, unstructuredReference=Chupp W, Zubrow A, Raymond A, et al. MOVES3 sensitivity analysis[R]. Washington D C: Federal Highway Administration, 2023., articleTitle=MOVES3 sensitivity analysis, refAbstract=null), Reference(id=1251595997613339200, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2010, volume=44, issue=25, pageStart=2943, pageEnd=2953, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=Smit R, Ntziachristos L, Boulter P, journalName=Atmospheric Environment, refType=null, unstructuredReference=Smit R, Ntziachristos L, Boulter P. Validation of road vehicle and traffic emission models: a review and Meta-analysis[J]. Atmospheric Environment, 2010, 44(25): 2943-2953., articleTitle=Validation of road vehicle and traffic emission models: a review and Meta-analysis, refAbstract=null), Reference(id=1251595997684642370, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2024, volume=130, issue=null, pageStart=104157, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=Fan P, Song G, Zhai Z, journalName=Transportation Research Part D: Transport and Environment, refType=null, unstructuredReference=Fan P, Song G, Zhai Z, et al. Fuel consumption estimation in heavy-duty trucks: integrating vehicle weight into deep-learning frameworks[J]. Transportation Research Part D: Transport and Environment, 2024, 130: 104157., articleTitle=Fuel consumption estimation in heavy-duty trucks: integrating vehicle weight into deep-learning frameworks, refAbstract=null), Reference(id=1251595997743362628, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2017, volume=17, issue=12, pageStart=108, pageEnd=114, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=阳冬波, 王敏, 邱兆文, journalName=科学技术与工程, refType=null, unstructuredReference=阳冬波, 王敏, 邱兆文. 不同公路等级的汽车油耗和排放评估研究[J]. 科学技术与工程, 2017, 17(12): 108-114., articleTitle=不同公路等级的汽车油耗和排放评估研究, refAbstract=null), Reference(id=1251595997797888582, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2017, volume=17, issue=12, pageStart=108, pageEnd=114, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=Yang Dongbo, Wang Min, Qiu Zhaowen, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Yang Dongbo, Wang Min, Qiu Zhaowen. Research on evaluation of vehicle fuel consumption and emission of different highway classes[J]. Science Technology and Engineering, 2017, 17(12): 108-114., articleTitle=Research on evaluation of vehicle fuel consumption and emission of different highway classes, refAbstract=null), Reference(id=1251595997890163272, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2024, volume=16, issue=11, pageStart=4487, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=Gavric S, Erdagi I G, Stevanovic A, journalName=Sustainability, refType=null, unstructuredReference=Gavric S, Erdagi I G, Stevanovic A. Environmental assessment of incorrect automated pedestrian detection and common pedestrian timing treatments at signalized intersections[J]. Sustainability, 2024, 16(11): 4487., articleTitle=Environmental assessment of incorrect automated pedestrian detection and common pedestrian timing treatments at signalized intersections, refAbstract=null), Reference(id=1251595997990826570, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2024, volume=130, issue=null, pageStart=104157, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=Fan P, Song G, Zhai Z, journalName=Transportation Research Part D: Transport and Environment, refType=null, unstructuredReference=Fan P, Song G, Zhai Z, et al. Fuel consumption estimation in heavy-duty trucks: integrating vehicle weight into deep-learning frameworks[J]. Transportation Research Part D: Transport and Environment, 2024, 130: 104157., articleTitle=Fuel consumption estimation in heavy-duty trucks: integrating vehicle weight into deep-learning frameworks, refAbstract=null), Reference(id=1251595998062129740, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2024, volume=16, issue=5, pageStart=452, pageEnd=460, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=Ashqer M I, Ashqar H I, Elhenawy M, journalName=Transportation Letters, refType=null, unstructuredReference=Ashqer M I, Ashqar H I, Elhenawy M, et al. Evaluating a signalized intersection performance using unmanned aerial Data[J]. Transportation Letters, 2024, 16(5): 452-460., articleTitle=Evaluating a signalized intersection performance using unmanned aerial Data, refAbstract=null), Reference(id=1251595998125044302, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2020, volume=88, issue=11, pageStart=102576, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=Moradi E, Miranda-Moreno L, journalName=Transportation Research Part D: Transport and Environment, refType=null, unstructuredReference=Moradi E, Miranda-Moreno L. Vehicular fuel consumption estimation using real-world measures through cascaded machine learning modeling[J]. Transportation Research Part D: Transport and Environment, 2020, 88(11): 102576., articleTitle=Vehicular fuel consumption estimation using real-world measures through cascaded machine learning modeling, refAbstract=null), Reference(id=1251595998217318992, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2022, volume=45, issue=4, pageStart=129, pageEnd=132, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=张雨, journalName=中国航海, refType=null, unstructuredReference=张雨. 基于LASSO的船舶日油耗预测研究[J]. 中国航海, 2022, 45(4): 129-132, 139., articleTitle=基于LASSO的船舶日油耗预测研究, refAbstract=null), Reference(id=1251595998280233554, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2022, volume=45, issue=4, pageStart=129, pageEnd=132, url=null, language=null, rfNumber=[11], rfOrder=13, authorNames=Zhang Yu, journalName=Navigation of China, refType=null, unstructuredReference=Zhang Yu. Research on daily fuel consumption prediction of ships based on LASSO[J]. Navigation of China, 2022, 45(4): 129-132, 139., articleTitle=Research on daily fuel consumption prediction of ships based on LASSO, refAbstract=null), Reference(id=1251595998364119636, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2019, volume=74, issue=null, pageStart=174, pageEnd=188, url=null, language=null, rfNumber=[12], rfOrder=14, authorNames=Zargarnezhad S, Dashti R, Ahmadi R, journalName=Transportation Research Part D: Transport and Environment, refType=null, unstructuredReference=Zargarnezhad S, Dashti R, Ahmadi R. Predicting vehicle fuel consumption in energy distribution companies using ANNs[J]. Transportation Research Part D: Transport and Environment, 2019, 74: 174-188., articleTitle=Predicting vehicle fuel consumption in energy distribution companies using ANNs, refAbstract=null), Reference(id=1251595998443811414, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=151, issue=null, pageStart=105101, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=15, authorNames=Ghasemi M, Qazani M R C, Lennon C W, journalName=Journal of the Taiwan Institute of Chemical Engineers, refType=null, unstructuredReference=Ghasemi M, Qazani M R C, Lennon C W, et al. Analysis and prediction of microbial fuel cell behaviour using MLP and SVR[J]. Journal of the Taiwan Institute of Chemical Engineers, 2023, 151: 105101., articleTitle=Analysis and prediction of microbial fuel cell behaviour using MLP and SVR, refAbstract=null), Reference(id=1251595998510920280, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=149, issue=5, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=16, authorNames=Wang G, Zhang L, Xu Z, journalName=Part A: Systems, refType=null, unstructuredReference=Wang G, Zhang L, Xu Z, et al. Predictability of vehicle fuel consumption using LSTM: findings from field experiments[J]. Journal of Transportation Engineering, Part A: Systems, 2023, 149(5): 04023030., articleTitle=Predictability of vehicle fuel consumption using LSTM: findings from field experiments, refAbstract=null), Reference(id=1251595998573834842, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2015, volume=36, issue=5, pageStart=107, pageEnd=120, url=null, language=null, rfNumber=[15], rfOrder=17, authorNames=Walnum H J, Simonsen M, journalName=Transportation Research Part D, refType=null, unstructuredReference=Walnum H J, Simonsen M. Does driving behavior matter? an analysis of fuel consumption data from heavy-duty trucks[J]. Transportation Research Part D, 2015, 36(5): 107-120., articleTitle=Does driving behavior matter? an analysis of fuel consumption data from heavy-duty trucks, refAbstract=null), Reference(id=1251595998632555100, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2021, volume=12, issue=11, pageStart=406, pageEnd=414, url=null, language=null, rfNumber=[16], rfOrder=18, authorNames=Hamed M A, Khafagy M H, Badry R M, journalName=International Journal of Advanced Computer Science and Applications, refType=null, unstructuredReference=Hamed M A, Khafagy M H, Badry R M. Fuel consumption prediction model using machine learning[J]. International Journal of Advanced Computer Science and Applications, 2021, 12(11): 406-414., articleTitle=Fuel consumption prediction model using machine learning, refAbstract=null), Reference(id=1251595998695469662, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2017, volume=59, issue=null, pageStart=24, pageEnd=29, url=null, language=null, rfNumber=[17], rfOrder=19, authorNames=Du Y, Wu J, Yang S, journalName=Journal of Environmental Sciences, refType=null, unstructuredReference=Du Y, Wu J, Yang S, et al. Predicting vehicle fuel consumption patterns using floating vehicle data[J]. Journal of Environmental Sciences, 2017, 59: 24-29., articleTitle=Predicting vehicle fuel consumption patterns using floating vehicle data, refAbstract=null), Reference(id=1251595998758384223, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2018, volume=33, issue=3, pageStart=209, pageEnd=219, url=null, language=null, rfNumber=[18], rfOrder=20, authorNames=Xu Z, Wei T, Easa S, journalName=Computer-Aided Civil and Infrastructure Engineering, refType=null, unstructuredReference=Xu Z, Wei T, Easa S, et al. Modeling relationship between truck fuel consumption and driving behavior using data from internet of vehicles[J]. Computer-Aided Civil and Infrastructure Engineering, 2018, 33(3): 209-219., articleTitle=Modeling relationship between truck fuel consumption and driving behavior using data from internet of vehicles, refAbstract=null), Reference(id=1251595998825493089, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2021, volume=285, issue=null, pageStart=119137, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=21, authorNames=Wang J, Shen L, Bi Y, journalName=Fuel, refType=null, unstructuredReference=Wang J, Shen L, Bi Y, et al. Modeling and optimization of a light-duty diesel engine at high altitude with a support vector machine and a genetic algorithm[J]. Fuel, 2021, 285: 119137., articleTitle=Modeling and optimization of a light-duty diesel engine at high altitude with a support vector machine and a genetic algorithm, refAbstract=null), Reference(id=1251595998913573475, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=3, pageStart=121, pageEnd=126, url=null, language=null, rfNumber=[20], rfOrder=22, authorNames=王一婷, 行本贝, 李彬, journalName=计算机与现代化, refType=null, unstructuredReference=王一婷, 行本贝, 李彬, . 基于SHARED-LSTM的重型卡车不同加速驾驶行为油耗预测方法[J]. 计算机与现代化, 2023(3): 121-126., articleTitle=基于SHARED-LSTM的重型卡车不同加速驾驶行为油耗预测方法, refAbstract=null), Reference(id=1251595998976488037, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=3, pageStart=121, pageEnd=126, url=null, language=null, rfNumber=[20], rfOrder=23, authorNames=Wang Yiting, Hang Benbei, Li Bin, journalName=Computer and Modernization, refType=null, unstructuredReference=Wang Yiting, Hang Benbei, Li Bin, et al. Fuel consumption prediction method of heavy trucks with different accelerating driving behaviors based on SHARED-LSTM[J]. Computer and Modernization, 2023(3): 121-126., articleTitle=Fuel consumption prediction method of heavy trucks with different accelerating driving behaviors based on SHARED-LSTM, refAbstract=null), Reference(id=1251595999051985511, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=36, pageStart=16067, pageEnd=16074, url=null, language=null, rfNumber=[21], rfOrder=24, authorNames=王愈轩, 刘尔佳, 黄永章, journalName=科学技术与工程, refType=null, unstructuredReference=王愈轩, 刘尔佳, 黄永章. 基于CNN-LSTM-lightGBM组合的超短期风电功率预测方法[J]. 科学技术与工程, 2022, 22(36): 16067-16074., articleTitle=基于CNN-LSTM-lightGBM组合的超短期风电功率预测方法, refAbstract=null), Reference(id=1251595999119094377, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2019, volume=22, issue=36, pageStart=16067, pageEnd=16074, url=null, language=null, rfNumber=[21], rfOrder=25, authorNames=Wang Yuxuan, Liu Erjia, Huang Yongzhang, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Yuxuan, Liu Erjia, Huang Yongzhang. Ultra-short term wind power forecasting method based on CNN-LSTM-lightGBM combination[J]. Science Technology and Engineering, 2019, 22(36): 16067-16074., articleTitle=Ultra-short term wind power forecasting method based on CNN-LSTM-lightGBM combination, refAbstract=null), Reference(id=1251595999194591850, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=35, pageStart=15091, pageEnd=15099, url=null, language=null, rfNumber=[22], rfOrder=26, authorNames=程杰, 陈鼎, 李春, journalName=科学技术与工程, refType=null, unstructuredReference=程杰, 陈鼎, 李春, . 基于GWO-CNN-BiLSTM的超短期风电预测[J]. 科学技术与工程, 2023, 23(35): 15091-15099., articleTitle=基于GWO-CNN-BiLSTM的超短期风电预测, refAbstract=null), Reference(id=1251595999253312108, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=35, pageStart=15091, pageEnd=15099, url=null, language=null, rfNumber=[22], rfOrder=27, authorNames=Cheng Jie, Chen Ding, Li Chun, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Cheng Jie, Chen Ding, Li Chun, et al. Ultra-short term wind power prediction based on GWO-CNN-BiLSTM[J]. Science Technology and Engineering, 2023, 23(35): 15091-15099., articleTitle=Ultra-short term wind power prediction based on GWO-CNN-BiLSTM, refAbstract=null), Reference(id=1251595999324615278, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, doi=null, pmid=null, pmcid=null, year=2018, volume=18, issue=185, pageStart=1, pageEnd=52, url=null, language=null, rfNumber=[23], rfOrder=28, authorNames=Li L, Jamieson K, DeSalvo G, journalName=Journal of Machine Learning Research, refType=null, unstructuredReference=Li L, Jamieson K, DeSalvo G, et al. Hyperband: a novel bandit-based approach to hyperparameter optimization[J]. Journal of Machine Learning Research, 2018, 18(185): 1-52., articleTitle=Hyperband: a novel bandit-based approach to hyperparameter optimization, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1251595993297400254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, xref=null, ext=[AuthorCompanyExt(id=1251595993305788863, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China), AuthorCompanyExt(id=1251595993314177472, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, companyId=1251595993297400254, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=新疆农业大学交通与物流工程学院, 乌鲁木齐 830052)])], figs=[ArticleFig(id=1251595995130311165, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.1, caption=Thermal map of correlation between fuel consumption and its influencing factors, figureFileSmall=8fIccfRmrIrD3ISCMvbDVA==, figureFileBig=FQd8WeHkeekHk7izRiTLjg==, tableContent=null), ArticleFig(id=1251595995184837119, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图1, caption=油耗与其影响因子相关性热力图, figureFileSmall=8fIccfRmrIrD3ISCMvbDVA==, figureFileBig=FQd8WeHkeekHk7izRiTLjg==, tableContent=null), ArticleFig(id=1251595995268723202, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.2, caption=CNN structure diagram, figureFileSmall=hwy4FoPMS66EIs6pbyrbMw==, figureFileBig=HSKc0DeM9Lt/5Qn1Y9pbGg==, tableContent=null), ArticleFig(id=1251595995327443460, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图2, caption=CNN结构示意图, figureFileSmall=hwy4FoPMS66EIs6pbyrbMw==, figureFileBig=HSKc0DeM9Lt/5Qn1Y9pbGg==, tableContent=null), ArticleFig(id=1251595995402940934, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.3, caption=LSTM network structure diagram, figureFileSmall=sItEmkPCZ3hPe8LYwxfdtQ==, figureFileBig=Qz1K46m3I3Z2Mjnfl1RB8Q==, tableContent=null), ArticleFig(id=1251595995461661191, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图3, caption=LSTM网络结构图, figureFileSmall=sItEmkPCZ3hPe8LYwxfdtQ==, figureFileBig=Qz1K46m3I3Z2Mjnfl1RB8Q==, tableContent=null), ArticleFig(id=1251595995532964361, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.4, caption=BiLSTM network structure diagram, figureFileSmall=Zis+S2s9dPjpf79TaB3Jqw==, figureFileBig=vWCNHfUscDN/+WENl6wU5A==, tableContent=null), ArticleFig(id=1251595995591684619, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图4, caption=BiLSTM网络结构图, figureFileSmall=Zis+S2s9dPjpf79TaB3Jqw==, figureFileBig=vWCNHfUscDN/+WENl6wU5A==, tableContent=null), ArticleFig(id=1251595995662987789, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.5, caption=CNN-BiLSTM model network structure, figureFileSmall=XtP189YCpDOVKp1XyCDFRA==, figureFileBig=NbAr1mGBaD9pwmsvd4durA==, tableContent=null), ArticleFig(id=1251595995730096655, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图5, caption=CNN-BiLSTM模型网络结构, figureFileSmall=XtP189YCpDOVKp1XyCDFRA==, figureFileBig=NbAr1mGBaD9pwmsvd4durA==, tableContent=null), ArticleFig(id=1251595995809788433, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.6, caption=Flow chart of CNN-BiLSTM fuel consumption prediction model, figureFileSmall=rW5RqzTSyYqyHmn9VSS1cA==, figureFileBig=/xWcwmEftOwgvU8E8uf2Lg==, tableContent=null), ArticleFig(id=1251595995902063123, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图6, caption=CNN-BiLSTM油耗预测模型流程图, figureFileSmall=rW5RqzTSyYqyHmn9VSS1cA==, figureFileBig=/xWcwmEftOwgvU8E8uf2Lg==, tableContent=null), ArticleFig(id=1251595995952394773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Fig.7, caption=A fitting curve of partial true and predicted values, figureFileSmall=Z46uEoFkyeT6LLFCx82Y/A==, figureFileBig=5vLyXf8GYgnG5xTj/RdeoA==, tableContent=null), ArticleFig(id=1251595996019503638, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=图7, caption=部分真实值和预测值的拟合曲线, figureFileSmall=Z46uEoFkyeT6LLFCx82Y/A==, figureFileBig=5vLyXf8GYgnG5xTj/RdeoA==, tableContent=null), ArticleFig(id=1251595996090806807, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 1, caption=

Some examples of vehicle driving status data

, figureFileSmall=null, figureFileBig=null, tableContent=
样本序号 车速/
(km·h-1)
加速度/
(m·s-2)
节气门
位置/%
负荷率/% 发动机转速/
(r·min-1)
冷却液
温度/℃
进气管绝对
压力/kPa
燃油消耗/
(g·s-1)
1 0 0 5.2 23.9 929 87 33 0.268 8
2 0 0 6.1 30.6 726 87 40 0.243 2
3 0 0 4.7 31.8 685 87 41 0.217 6
4 0 0 5.7 29 675 87 38 0.204 8
21 032 5 -1.388 8.0 52.2 893 89 62 0.486 4
), ArticleFig(id=1251595996174692889, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表1, caption=

部分车辆行驶状态数据示例

, figureFileSmall=null, figureFileBig=null, tableContent=
样本序号 车速/
(km·h-1)
加速度/
(m·s-2)
节气门
位置/%
负荷率/% 发动机转速/
(r·min-1)
冷却液
温度/℃
进气管绝对
压力/kPa
燃油消耗/
(g·s-1)
1 0 0 5.2 23.9 929 87 33 0.268 8
2 0 0 6.1 30.6 726 87 40 0.243 2
3 0 0 4.7 31.8 685 87 41 0.217 6
4 0 0 5.7 29 675 87 38 0.204 8
21 032 5 -1.388 8.0 52.2 893 89 62 0.486 4
), ArticleFig(id=1251595996250190363, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 2, caption=

Analysis of raw data statistical results

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 车速/
(km·h-1)
加速度/
(m·s-2)
节气门
位置/%
负荷率/% 发动机转速/
(r·min-1)
冷却液
温度/℃
进气管绝对
压力/kPa
燃油消耗/
(g·s-1)
数据总量 21 032 21 032 21 032 21 032 21 032 21 032 21 032 21 032
平均值 34.69 0.000 3 14.90 42.84 1 563.59 75.44 39.31 0.882 8
标准差 27.42 0.498 5 18.90 20.10 630.01 19.06 28.58 0.875 9
最大值 129 2.5 100 98.40 6 206 99 175 10.828 8
最小值 0 -4.166 7 4.70 11 617 12 0 0
上四分位数 9 -0.277 8 7.10 26.30 979 67 17 0.345 6
中分位数 30 0 8.50 40.40 1656 81 40 0.524 8
下四分位数 61 0.277 8 14.10 52.50 1979 90 58 1.228 8
), ArticleFig(id=1251595996325687837, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表2, caption=

原始数据统计分析结果

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 车速/
(km·h-1)
加速度/
(m·s-2)
节气门
位置/%
负荷率/% 发动机转速/
(r·min-1)
冷却液
温度/℃
进气管绝对
压力/kPa
燃油消耗/
(g·s-1)
数据总量 21 032 21 032 21 032 21 032 21 032 21 032 21 032 21 032
平均值 34.69 0.000 3 14.90 42.84 1 563.59 75.44 39.31 0.882 8
标准差 27.42 0.498 5 18.90 20.10 630.01 19.06 28.58 0.875 9
最大值 129 2.5 100 98.40 6 206 99 175 10.828 8
最小值 0 -4.166 7 4.70 11 617 12 0 0
上四分位数 9 -0.277 8 7.10 26.30 979 67 17 0.345 6
中分位数 30 0 8.50 40.40 1656 81 40 0.524 8
下四分位数 61 0.277 8 14.10 52.50 1979 90 58 1.228 8
), ArticleFig(id=1251595996380213791, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
Algorithm 1 Hyperband Algorithm
Input: Rmax,α %By Default: α=3
Output: Cbest %By Configuration
Initialize: mmax=logα(Rmax),
R=(mmax+1)${{R}_{\mathrm{m}\mathrm{a}\mathrm{x}}}_{.}$
for m∈{mmax,mmax-1,…,0}
n=$\frac{{R}_{\alpha }^{m}}{{R}_{\mathrm{m}\mathrm{a}\mathrm{x}}(m+1)}$, t = Rmaxα-8.
In this iteration (n,t) are the parameters
for Successive-Halving.
C = get_hyperparameter_config(n)
for i∈{0,1,2,…,m}
ni=-m
ti=i
VL= return_val_loss(c,ti), cC
C = get_K_config $\left(C,\mathrm{V}\mathrm{L},\frac{{n}_{i}}{\alpha }\right)$
end for
end for
return Configuration with least validation loss Cbest
), ArticleFig(id=1251595996438934049, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=, caption=

, figureFileSmall=null, figureFileBig=null, tableContent=
Algorithm 1 Hyperband Algorithm
Input: Rmax,α %By Default: α=3
Output: Cbest %By Configuration
Initialize: mmax=logα(Rmax),
R=(mmax+1)${{R}_{\mathrm{m}\mathrm{a}\mathrm{x}}}_{.}$
for m∈{mmax,mmax-1,…,0}
n=$\frac{{R}_{\alpha }^{m}}{{R}_{\mathrm{m}\mathrm{a}\mathrm{x}}(m+1)}$, t = Rmaxα-8.
In this iteration (n,t) are the parameters
for Successive-Halving.
C = get_hyperparameter_config(n)
for i∈{0,1,2,…,m}
ni=-m
ti=i
VL= return_val_loss(c,ti), cC
C = get_K_config $\left(C,\mathrm{V}\mathrm{L},\frac{{n}_{i}}{\alpha }\right)$
end for
end for
return Configuration with least validation loss Cbest
), ArticleFig(id=1251595996510237219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 3, caption=

Examples of data normalization processing results

, figureFileSmall=null, figureFileBig=null, tableContent=
速度/
(km·h-1)
加速度/
(m·s-2)
节气门
位置%
负荷率% 发动机转速/
(r·min-1)
冷却液
温度/℃
进气管绝对
压力/kPa
燃油消耗/
(g·s-1)
0 0 0.005 2 0.135 7 0.055 4 0.820 8 0.185 1 0.268 8
0 0 0.014 6 0.213 4 0.019 1 0.820 8 0.271 6 0.243 2
0 0 0.005 2 0.227 3 0.011 8 0.820 8 0.283 9 0.217 6
0 0 0.010 49 0.194 8 0.010 0 0.820 8 0.246 9 0.204 8
0 0 0.010 49 0.208 8 0.017 3 0.820 8 0.271 6 0.217 6
), ArticleFig(id=1251595996581540389, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表3, caption=

数据归一化处理结果部分示例

, figureFileSmall=null, figureFileBig=null, tableContent=
速度/
(km·h-1)
加速度/
(m·s-2)
节气门
位置%
负荷率% 发动机转速/
(r·min-1)
冷却液
温度/℃
进气管绝对
压力/kPa
燃油消耗/
(g·s-1)
0 0 0.005 2 0.135 7 0.055 4 0.820 8 0.185 1 0.268 8
0 0 0.014 6 0.213 4 0.019 1 0.820 8 0.271 6 0.243 2
0 0 0.005 2 0.227 3 0.011 8 0.820 8 0.283 9 0.217 6
0 0 0.010 49 0.194 8 0.010 0 0.820 8 0.246 9 0.204 8
0 0 0.010 49 0.208 8 0.017 3 0.820 8 0.271 6 0.217 6
), ArticleFig(id=1251595996652843559, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 4, caption=

Model input hyperparameters

, figureFileSmall=null, figureFileBig=null, tableContent=
模型参数 参数选择
选择值 最大值 最小值
卷积核数 0 256
神经元个数 0 512
Dropout 0.1 0.5
学习率 0.1,0.01,0.001,0.000 1
), ArticleFig(id=1251595996715758121, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表4, caption=

模型输入超参

, figureFileSmall=null, figureFileBig=null, tableContent=
模型参数 参数选择
选择值 最大值 最小值
卷积核数 0 256
神经元个数 0 512
Dropout 0.1 0.5
学习率 0.1,0.01,0.001,0.000 1
), ArticleFig(id=1251595996816421419, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 5, caption=

Parameters of different models

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 LSTM BiLSTM CNN CNN-
LSTM
CNN-
BiLSTM
卷积核数 64 64 64
卷积核大小 1×2 1×2 1×2
池化层窗口 2 2 2
神经元个数 128 128 128 128 128
Dropout 0.5 0.5 0.5 0.5 0.5
), ArticleFig(id=1251595996900307501, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表5, caption=

不同模型的参数

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 LSTM BiLSTM CNN CNN-
LSTM
CNN-
BiLSTM
卷积核数 64 64 64
卷积核大小 1×2 1×2 1×2
池化层窗口 2 2 2
神经元个数 128 128 128 128 128
Dropout 0.5 0.5 0.5 0.5 0.5
), ArticleFig(id=1251595996967416367, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 6, caption=

Evaluation indicators of different models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型类别 评价指标
MAE RMSE R2
CNN 0.146 38 0.250 96 0.959 49
LSTM 0.078 87 0.221 02 0.962 46
BiLSTM 0.067 46 0.155 21 0.984 08
CNN-LSTM 0.076 25 0.152 93 0.986 45
CNN-BiLSTM 0.066 81 0.153 81 0.987 56
Hyperband-CNN-BiLSTM 0.057 69 0.119 25 0.991 76
), ArticleFig(id=1251595997055496753, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表6, caption=

不同模型的评价指标

, figureFileSmall=null, figureFileBig=null, tableContent=
模型类别 评价指标
MAE RMSE R2
CNN 0.146 38 0.250 96 0.959 49
LSTM 0.078 87 0.221 02 0.962 46
BiLSTM 0.067 46 0.155 21 0.984 08
CNN-LSTM 0.076 25 0.152 93 0.986 45
CNN-BiLSTM 0.066 81 0.153 81 0.987 56
Hyperband-CNN-BiLSTM 0.057 69 0.119 25 0.991 76
), ArticleFig(id=1251595997126799923, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=EN, label=Table 7, caption=

Evaluation indexes of the model under different time Windows

, figureFileSmall=null, figureFileBig=null, tableContent=
输入序列 MAE RMSE R2
1 0.075 71 0.163 15 0.958 68
2 0.057 69 0.119 25 0.991 76
3 0.064 00 0.142 15 0.988 77
4 0.057 71 0.127 55 0.990 25
5 0.059 25 0.134 38 0.989 01
6 0.060 00 0.132 36 0.990 18
), ArticleFig(id=1251595997189714485, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781955945919116, language=CN, label=表7, caption=

不同时间窗口下模型的评价指标

, figureFileSmall=null, figureFileBig=null, tableContent=
输入序列 MAE RMSE R2
1 0.075 71 0.163 15 0.958 68
2 0.057 69 0.119 25 0.991 76
3 0.064 00 0.142 15 0.988 77
4 0.057 71 0.127 55 0.990 25
5 0.059 25 0.134 38 0.989 01
6 0.060 00 0.132 36 0.990 18
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, 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=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, 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=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403249, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2403249, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2403249, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2403249, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于Hyperband-CNN-BiLSTM模型的车辆油耗预测方法
收藏切换
PDF下载
吐尔逊·买买提 , 孙慧 , 刘亚楼
科学技术与工程 | 论文·交通运输 2025,25(9): 3896-3904
收起
收藏切换
科学技术与工程 | 论文·交通运输 2025, 25(9): 3896-3904
基于Hyperband-CNN-BiLSTM模型的车辆油耗预测方法
全屏
吐尔逊·买买提 , 孙慧, 刘亚楼
作者信息
  • 新疆农业大学交通与物流工程学院, 乌鲁木齐 830052
  • 吐尔逊·买买提(1978—),男,维吾尔族,新疆阿克苏人,博士,副教授。研究方向:交通环境、数据挖掘。E-mail:

Vehicle Fuel Consumption Prediction Method Based on Hyperband-CNN-BiLSTM Model
Mamaiti TURSON , Hui SUN, Ya-lou LIU
Affiliations
  • College of Transportation and Logistics Engineering, Xinjiang Agricultural University, Urumqi 830052, China
出版时间: 2025-03-28 doi: 10.12404/j.issn.1671-1815.2403249
文章导航
收藏切换

为了有效地预测车辆的燃油消耗,提高燃油经济性并推动节能减排,提出一种基于Hyperband-CNN-BiLSTM的机动车油耗预测方法。首先基于实际道路测试收集到的车辆运行状态数据和油耗数据,分析了影响车辆油耗的显著性因素;其次结合卷积神经网络(convolutional neural network,CNN)强大的特征提取能力和双向长短期记忆网络(bidirectional long short-term memory,BiLSTM)在处理时序数据方面的优势,构建了基于CNN-BiLSTM的车辆油耗预测组合模型;然后,为提高模型预测准确性,通过Hyperband优化算法对组合模型进行优化,并将车辆油耗影响因素作为模型输入特征,对模型进行训练,实现对车辆油耗的建模和预测;最后,选取CNN、LSTM、BiLSTM、CNN-LSTM、CNN-BiLSTM作为对比模型,对Hyperband-CNN-BiLSTM预测模型效果进行评价。结果表明,相较于其他模型,Hyperband-CNN-BiLSTM模型的平均绝对误差(mean absolute error,MAE)和均方根误差(root mean squared error,RMSE)最小,分别为0.057 69和0.119 25,R2最大,为0.991 76,模型预测效果最佳。

Hyperband  /  油耗预测  /  卷积神经网络(CNN)  /  双向长短期记忆网络(BiLSTM)  /  组合模型

In order to effectively predict the fuel consumption of vehicles, improve fuel economy and promote energy saving and emission reduction, a Hyperband-CNN-BiLSTM-based motor vehicle fuel consumption prediction method was proposed. Firstly, based on the vehicle operating status data and fuel consumption data collected from the actual road test, the salient factors affecting the fuel consumption of vehicles were analyzed. Secondly, combining the powerful feature extraction capability of convolutional neural network(CNN) and the advantages of bidirectional long and short-term memory network (BiLSTM) in dealing with the time-series data, a combined model of vehicle fuel consumption prediction based on CNN-BiLSTM was constructed. Then, in order to improve the model prediction accuracy, the combined model was optimized by Hyperband optimization algorithm, and the vehicle fuel consumption influencing factors were taken as the model input features to train the model to realize the modeling and prediction of vehicle fuel consumption. Finally, CNN, LSTM, BiLSTM, CNN-LSTM and CNN-BILSTM were selected as comparison models to evaluate the effect of Hyperband-CNN-BiLSTM prediction model. The results show that compared with other models, the Hyperband-CNN-BiLSTM model has the smallest mean absolute error (MAE) and root mean squared error (RMSE). They are 0.057 69 and 0.119 25, respectively. R2 is the largest (0.991 76), and the model has the best prediction effect.

Hyperband  /  fuel rate prediction  /  convolutional neural networks (CNN)  /  bidirectional long short-term memory network (BiLSTM)  /  combination model
吐尔逊·买买提, 孙慧, 刘亚楼. 基于Hyperband-CNN-BiLSTM模型的车辆油耗预测方法. 科学技术与工程, 2025 , 25 (9) : 3896 -3904 . DOI: 10.12404/j.issn.1671-1815.2403249
Mamaiti TURSON, Hui SUN, Ya-lou LIU. Vehicle Fuel Consumption Prediction Method Based on Hyperband-CNN-BiLSTM Model[J]. Science Technology and Engineering, 2025 , 25 (9) : 3896 -3904 . DOI: 10.12404/j.issn.1671-1815.2403249
随着国家经济的蓬勃发展和工业化进程的加速推进,中国机动车数量持续攀升,交通运输业也呈现出迅猛的发展态势。这一变化虽为人们的日常生活带来了诸多便利,但也引发了能源短缺和环境污染等一系列负面问题。生态环境部2023年12月7日发布的《中国移动源环境管理年报(2023年)》[1]显示,车辆的燃油排放已成为中国大中城市空气污染的重要来源,因此,提升燃油经济性、推进节能减排成为国内外学者共同关注的焦点,车辆燃油消耗模型的研究也日益受到重视。目前,关于车辆油耗预测的研究主要沿着两大方向进行。一方面,研究者们深入探索基于车辆动力学原理的传统油耗模型;另一方面,随着大数据和机器学习技术的兴起,研究者们也开始积极尝试基于机器学习方法的数据驱动油耗模型。其中传统油耗模型以弗吉尼亚理工大学综合动力燃料消耗模型[2](virginia tech comprehensive power-based fuel consumption model, VT-CPFM)、综合移动源排放模型[3](motor vehicle emission simulator,MOVES)、微观尾气排放计算模型[4](comprehensive modal emission model, CMEM)和微观排放模型[5](virginia tech microscopic, VT-Micro)等为代表,这些模型从宏观、中观、微观3个不同层次对交通燃油排放及其演变趋势进行了全面而深入的预测。MOVES模型[6]是美国环保署研发的新一代综合性排放模型,综合考虑了车辆燃油类型、行驶里程、车龄以及环境因素等多方面因素。通过分析车辆比功率的分布特性,MOVES模型能够精确描绘不同运行工况下车辆的燃油消耗和污染排放特征,具有一定的代表性。然而,MOVES模型对输入数据的庞大需求和复杂的计算过程,使其难以实现油耗的实时估算和即时反馈,在一定程度上限制了其在实际应用中的实时性和便捷性。CMEM模型[7]和VT-Micro模型[8]则通过建立车辆动力学参数和车辆燃油耗之间的多项式函数关系,实现了对车辆瞬时燃油耗的预测,但这两个模型对车辆运行状态数据要求较高,数据采集难度大,一定程度上限制了模型的应用范围。VT-CPFM模型[9]是一种基于车辆功率的微观燃油消耗模型,该模型利用二次多项式模型对复杂的燃油消耗与车辆功率之间的关系进行简化,从而实现对车辆瞬时油耗的预测,然而,由于模型在开发时仅基于特定车辆和驾驶条件进行校准,而实际行驶中车辆油耗受多种复杂因素的影响,影响模型的计算精度。
相较于传统动力学油耗模型,基于数据驱动的油耗模型通过深入挖掘车辆运行状态数据,构建状态参数与油耗之间的复杂映射关系,无需对模型的物理含义进行精确阐述,因此具有更高的预测精确度、更好的适用性和实时性。目前,油耗预测的主流数据模型包括支持向量回归[10](support vector regression, SVR)、套索算法[11](least absolute shrinkage and selection operator, LASSO)等机器学习模型,和人工神经网络[12](artificial neural networks, ANN)、多层感知机[13](multilayer perceptron, MLP)、长短期记忆网络[14](long short-term memory, LSTM)等深度学习模型。Walnum等[15]采用多元回归方法构建了卡车燃油消耗模型,并分析了各项驾驶指标与卡车燃油消耗之间的关系,揭示了对卡车油耗产生显著影响的关键因素。然而,多元回归方法在处理变量间复杂关系时存在一定的局限性,这在一定程度上影响了模型的预测精度。Hamed等[16]基于支持向量机(support vector machine, SVM)探究了车辆油耗与车辆动力学参数之间的关系,实现了对车辆油耗的预测。然而,SVM模型在处理变量间的线性关系不明显的非线性回归问题时存在一定的局限性,这导致其预测结果的准确性下降。Du等[17]基于大量浮动车历史数据,分析了油耗与多种影响因素之间的关系,并使用BP(back propagation)神经网络建立了油耗预测模型。但BP神经网络难以有效捕捉时序数据中的长期依赖关系,且大规模数据集中往往夹杂着一定数量的冗余信息,在一定程度上削弱了模型的准确性和泛化性。Xu等[18]基于车联网的动态油耗数据,使用广义回归神经网络建立了驾驶行为与油耗之间的关系模型,实现了对车辆油耗的预测,但车联网数据上传容易受到网络延迟影响,并且传统的RNN网络容易出现梯度消失和梯度爆炸等问题,影响模型的鲁棒性。Wang等[19]利用遗传算法(genetic algorithm, GA)优化SVM中的超参数,构建了GA-SVM车辆油耗预测模型,实现车辆油耗的准确预测。GA算法在优化过程中能够一定程度上提升SVM的性能,但在面对高维复杂问题时,GA算法容易陷入局部最优解,难以收敛到全局最优解。王一婷等[20]对车辆不同加速驾驶行为与油耗之间的关系进行了定量分析,并基于一种共享权重的LSTM模型建立了车辆油耗预测模型,实现了车辆油耗较为精确的预测。然而,该模型的超参数采用经验法确定,在一定程度上影响了模型的预测精度。
可以发现,基于数据驱动的油耗模型在应用时,常因大样本数据的特征冗余和模型超参数选取不当而导致预测精度受损。因此,为了提高模型预测精度,实现车辆油耗的精确预测,现首先分析影响车辆油耗的显著性因素。随后将Hyperband、CNN、BiLSTM相结合,构建Hyperband-CNN-BiLSTM模型。其中,CNN模型能有效提取模型输入特征,增强模型的鲁棒性和泛化性;BiLSTM模型则能捕捉数据间的长时序关系,一定程度上避免梯度爆炸和梯度消失的问题;而Hyperband则用于优化CNN-BiLSTM神经网络中的超参数,从而强化模型性能,提高预测精度。最后使用实际算例验证所提出模型的有效性,以期为机动车油耗预测问题提供一种有效的解决方法。
使用的数据样本基于实际道路驾驶测试,选取一名经验丰富的驾驶员,驾驶车辆沿乌鲁木齐克拉玛依西路、喀什东路和河滩快速路进行多次往复行驶,通过车载传感器实时获取的车辆行驶状态数据,车辆行驶工况包括了城市道路、市郊道路和快速路。原始数据集包含了行驶车速V,进气管绝对压力P,节气门位置TP,负荷率L,发动机转速n,冷却液温度CT和加速度a等关键参数。数据采集频率为1 Hz,共计收集了21 032条记录。表1展示了部分原始数据的样例。
对获取的原始数据中各参数做基本统计,数据基本统计结果如表2所示。可以发现,各参数总量均为21 032,无空缺值,无异常值。但数据中各参数数据的量值范围差异过大,符号不统一等情况,需要在后续研究时进行归一化处理,以解决数据量纲不统一的问题。
在车辆行驶过程中,燃油消耗受到多种因素的影响,其中既包括车辆的运行状态,也涉及发动机的工作状态。为了提取出对油耗产生显著影响的因素,对数据集进行了相关性矩阵的可视化处理,如图1所示。
图1可知,发动机转速、进气管压力、节气门位置和负载率与油耗的相关性系数均大于0.5,具有强相关性;速度和加速度与油耗的相关性系数介于0.2~0.5,具有一般相关性,而冷却液温度与油耗相关系数小于0.2,相关性较弱。因此,选择上述7个影响因素作为模型输入特征变量。
卷积神经网络CNN是一种前馈神经网络模型,够对数据进行高层次和抽象化的处理,其内部采用权值共享和局部连接的结构,可以在降低模型复杂度的同时,有效的提取数据中蕴含的深层次信息。CNN模型由卷积层、池化层、全连接层等结构组成。一维CNN结构如图2所示。
其中,卷积层通过卷积运算,提取数据的局部特征;随后,池化层对卷积层输出的特征矩阵进行重采样,降低特征维度,增强模型的鲁棒性;最后,全连接层将池化层输出的特征向量进行整合,完成回归任务。假设H为输入矩阵,I为核矩阵,mn分别为结果矩阵R的行列索引,卷积运算可用表示为
R[m,n]=(HI)[m,n]=$\sum _{j}\sum _{k}$I[j,k]H[m-j,n-k]
CNN因其强大的特征提取能力,常被用于处理时间序列、图像、音频等数据,其中一维卷积神经网络适用于处理时间序列数据。能够有效捕捉数据中的局部依赖和隐藏关系。由于原始车辆油耗数据是一维时间序列数据,故使用一维卷积神经网络提取油耗数据集中的特征信息。
长短期记忆网络(LSTM)是循环神经网络的一种特殊类型,通过引入门控机制解决了传统递归神经网络(recurrent neural network,RNN)在处理长时序数据时出现的梯度消失和梯度爆炸问题[21]。如图3所示,每个LSTM单元由1个记忆单元和3个门组成,这些门通过决定哪些信息将被遗忘,哪些信息将留在模型中来调节信息流,这3个门分别为输入门、遗忘门和输出门。遗忘门通过对当前输入Xt和前一个隐藏状态(ht-1)中包含的信息使用sigmoid函数(函数记为ft),并返回一个在(0,1)的值,这个值表示保存信息的百分比。输出门从当前输入和前一个隐藏状态获取信息,并通过第二个sigmoid函数转为(0,1)的值,同样的信息通过tanh函数,返回一个在(-1,1)的值。然后用sigmoid的输出it和tanh输出Ct的积来决定信息的重要性。然后将前一个神经单元状态(Ci-1)乘以遗忘门输出,加上输入门的输出,形成新的神经单元状态(Ct)。输出门决定下一个隐藏状态的值,来自当前输入和先前隐藏状态的信息通过第3个sigmoid函数传递,,新的神经单元状态通过tanh函数传递,将两个输出逐点相乘,输出门的输出(ot)将是新的隐藏状态(ht)。此时新的状态和新的隐藏状态被带到下一个时间步。LSTM的状态更新过程和最终输出过程可以简写为

ft=σ(Wf[ht-1,xt]+bf)

it=σ(Wi[ht-1,xt]+bi)

Ct=tanh(Wc[ht-1,xt]+bc)

Ct+1=ftCt-1+itCt

Ot=σ(Wo[ht-1,xt],bo)

ht=Ottanh(Ct)

式中:σ为sigmiod函数;xt为时间t的输入数据;ht为时间t的隐藏状态,每个W表示一个权值矩阵;每个b代表一个偏置向量。
BiLSTM网络包括前向LSTM和后向LSTM网络。如图4所示,前向LSTM使用t-k~t的值作为输入,而后向LSTM使用tt-k的值作为输入,前向$\overrightarrow{{h}_{t}}$和后向$\overleftarrow{{h}_{t}}$网络的输出通过使用之前描述的单个LSTM的机制来计算,则BiLSTM的输出(Yt)表达式为
Yt=σ($\overrightarrow{{h}_{t},}\overleftarrow{{h}_{t}}$)
CNN通过卷积层和池化层的交替叠加,能够高效的提取数据中的局部特征,进而捕捉到更深层次的信息。BiLSTM以其独特的选择性学习能力,有效解决了单向训练导致的信息利用不足问题,使得模型在处理时序数据时更具优势[22]。因此,通过串联CNN模型和BiLSTM模型,构建了一个CNN-BiLSTM混合神经网络。具体来说,模型包含CNN模块和BiLSTM模块。CNN模块由一个卷积层、一个池化层和一个Flatten层组成。负责提取输入数据的局部特征,并将其转化为BiLSTM模块的输入。第二个模块包括两个BiLSTM层、一个dropout层和一个dense层,BiLSTM层用于捕捉双向的时序信息,dropout层用于防止过拟合,提高模型的泛化能力,而dense层则负责整合特征信息并输出最终的油耗预测结果。其网络结构图如图5所示。
Hyperband优化算法是Successive-Halving优化算法的改进,有效解决了Successive-Halving算法中参数组合的资源分配问题[23]。该算法结合了随机搜索和动态资源分配的策略,可以在相对较短的时间内找到最优的超参数组合。Hyperband优化算法通过固定的总计算资源R(例如训练过程中最大迭代次数),实现了资源在不同迭代阶段n的高效分配策略。具体的,对于每个n值,都设定了一个最小资源或时间段t的分配规则,即n较小时,分配较多的资源t以促进充分探索;n增大时,则减少资源t以加快收敛速度。算法采用内外两层循环结构:内循环在固定资源约束下,针对当前的nt执行Successive-Halving过程,即对不同nt的组合执行连续减半运算(也称为backer);外循环负责迭代更新nt的值,引导算法逐步逼近最优解。Hyperband算法的的输入参数包括:单个迭代可分配的最大资源Rmax;每次循环中需要淘汰的参数比例α。这两个参数决定了连续减半运算的次数。同时,该算法根据初始资源确定初始n的值为mmax+1,即

m=mmax=logα(Rmax)

R=(mmax+1)Rmax

算法流程分为探索和利用两个阶段。在初始阶段,算法利用最大的减半次数mmax,在Rmax资源下广泛探索参数配置空间;随着迭代进行,n按一定系数τ递减,资源分配逐渐均匀,直到达到最小值n,此时所有的迭代都将获得相同的资源R。这种分阶段策略既保证了前期的广泛探索,又兼顾了后期的精确收敛。算法伪代码如下。
机动车油耗作为一种时间序列,受到道路状况和驾驶行为影响,具有一定的随机性和波动性,为了准确预测机动车燃油消耗,建立的基于Hyperband-CNN-BiLSTM油耗预测模型的流程图如图6所示。其在建立的CNN-BiLSTM模型基础上,将Hyperband优化模块添加到模型中,对CNN模块和BiLSTM模块进行优化,增强模型的特征提取和特征表现能力,增加数据相关性,使得模型能够学习油耗特征间更深层次的非线性关系,提高预测精度。
实验将车辆油耗设定为目标预测值,并将与之相关的各种参数作为特征值输入到预测模型中,以实现对车辆油耗的准确预测。结合表2可以发现,各个特征变量的量值范围存在显著差异,且符号的正负也不统一。为了确保数据在输入模型时具有统一的量纲,并消除因量值差异导致的预测偏差,对所有特征数据进行了归一化处理。归一化公式为

xα=(xi-xmin)/(xmax-xmin)

式(11)中:xα为特征值的归一化值;xi为实际特征值;xmax为特征数据中的最大值;xmin为特征数据中的最小值。
通过式(2),将样本数据归一化至(0,1),归一化后数据如表3所示。
实验环境配置如下:处理器为Intel(R) Core(TM) i5-4200H CPU @2.80GHz,GPU采用NVIDIA GeForce GTX950M,基于开源的Python环境管理平台Anaconda3,Python版本为3.10。实验使用深度学习框架Tensorflow中的Keras2.1.0版本构建所有神经网络模型。
Hyperband算法会按照输入的超参数生成最优参数组合,输入的超参数如表4所示。
为了验证模型的有效性,实验选择了4种神经网络模型与CNN-BiLSTM模型进行对比,4种模型分别为LSTM、BiLSTM、CNN、CNN-LSTM、CNN-BiLSTM。各模型参数设置如表5所示。
CNN层使用ReLU激活函数,其他网络层使用ReLU激活函数,各模型均使用改进自适应模型优化器Adam函数,损失函数为平均绝对误差,Epoch设置为100,学习率设为0.001,Batch Size设为2。
为了全面评估模型训练结果的准确性和性能,选取了3种评价指标: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)。这些指标的定义和计算公式如下。
(1)平均绝对误差(MAE)。MAE能够反映预测值与实际值间的误差大小。MAE的值越小,说明模型的预测准确度越高。计算公式为
MAE=$\frac{1}{m}\stackrel{m}{\sum _{i=1}}\left|({y}_{i}-{\stackrel{\wedge }{y}}_{i})\right|$
(2)均方根误差(RMSE)。RMSE用于衡量预测值与真实值之间的误差。RMSE的值越小,模型的准确度越高。计算公式为
RMSE=$\sqrt{\left[\frac{1}{m}\stackrel{m}{\sum _{i=1}}({y}_{i}-{\stackrel{\wedge }{y}}_{i}{)}^{2}\right]}$
(3)相关系数(R2)。R2用于反映模型对观测值的拟合程度。R2的取值越接近1,说明模型的拟合效果越好,即模型的鲁棒性越好。计算公式为
R2=1-$\frac{\sum _{i}({y}_{i}-{\stackrel{\wedge }{y}}_{i}{)}^{2}}{\sum _{i}(\overline{{y}_{i}}-{y}_{i}{)}^{2}}$
式中:yi$\stackrel{\wedge }{{y}_{i}}$分别为预测值和实际值;$\overline{{y}_{i}}$为实际值的平均值;m为数据数量。
将经过归一化处理的特征数据输入Hyperband-CNN-BiLSTM组合模型中,以预测车辆的油耗。随后,将得到的油耗预测值与真实值进行了详细比较。为了更直观地展示预测效果,从预测集与真实集中随机选取了连续的1 000条数据作为展示样本,并绘制了整体拟合曲线图,如图7所示。观察发现,所有模型整体上均展现出了较好的拟合效果。值得注意的是,在曲线波峰处,BiLSTM、CNN-LSTM、CNN-BiLSTM以及Hyperband-CNN-BiLSTM模型的表现尤为突出,这一点在图7上方的局部放大图中得到了更为清晰的展现。在曲线波谷处,CNN和LSTM模型稍显不足,出现了拟合不够准确的情况,如图7下方放大处所示。相比之下,Hyperband-CNN-BiLSTM模型在整个预测过程中表现稳定,无论是在波峰还是波谷处,都能够准确地跟随真实值的变化趋势,整体预测性能优异,进一步证明了其在汽车油耗预测中的有效性。
表6为不同模型的评价指标。可以明确看出,组合模型在MAE和MSE方面均显著优于单一模型,同时在R2上也呈现出显著的优势。这充分说明组合模型在整体性能上明显优于单一模型,具有更高的预测准确度和更好的数据拟合效果。进一步对CNN-LSTM、CNN-BiLSTM和Hyperband-CNN-BiLSTM这3个模型进行对比分析,发现后者在各项评价指标上均表现更为出色。具体而言,与CNN-LSTM模型和CNN-BiLSTM相比,Hyperband-CNN-BiLSTM模型的MAE相比前两者降低了24.3%、13.6%,RMSE降低了22%、22.4%,R2提高了0.53%和0.42%。这些数据说明Hyperband-CNN-BiLSTM模型在预测性能和拟合效果上的显著优势,充分验证了模型在车辆油耗预测中的有效性和优越性。
在进行时间序列预测时,选择合适的时间窗口序列对于Hyperband-CNN-BiLSTM模型的性能至关重要。选择了6种不同的时间窗口序列作为模型输入序列,模型的评价指标结果如表7所示。
经过多次试验发现,当模型的输入序列长度设定为2或4时,模型表现出最佳的预测结果。当输入序列长度为2时,预测结果相较于其他输入序列的预测结果,MAE分别降低了23.8%、9.86%、0.03%、2.63%、3.85%,RMSE分别降低了26.9%、16.1%、6.51%、11.26%、9.9%,R2分别提高了3.45%、0.3%、0.15%、0.28%、0.16%。可以发现,当输入的时间窗口序列较长时,会使模型出现过拟合现象,导致预测精度下降。
以实际道路测试所收集的车辆运行状态数据和油耗数据为基础,分析了影响车辆油耗的显著性因素。同时在确保数据时间序列的严密性的基础上,利用深度学习方法,提出了基于Hyperband-CNN-BiLSTM网络的机动车油耗预测模型,实现了油耗预测。并通过与其他模型进行对比,得出以下结论。
(1)所建立的CNN-BiLSTM模型通过一维CNN从局部到整体地挖掘了各参数量之间的关联关系,同时结合BiLSTM捕捉到了数据深层次的时间相关性,预测精度较CNN、LSTM、BiLSTM、CNN-LSTM模型分别提高了2.92%、2.61%、0.35%、0.11%。
(2)考虑到车辆油耗历史数据中不同特征对预测结果具有差异化的影响程度,通过引入Hyperband优化模型进行训练,有效提高了基础模型在处理非线性特征及深层次弱关联特征放方面的能力,使模型能更加精准的把握油耗特征与预测值之间的隐藏关系,提高了预测摩模型准确度。相较于未使用Hyperband优化的基础模型,预测精度提高了0.43%,实现了油耗的进一步优化和精准化。
参考文献 引证文献
排序方式:
[1]
中华人民共和国生态环境部. 中国移动源环境管理年报(2023年)[EB/OL]. (2023-12-07)[2024-05-02]. https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml. https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml
Ministry of Ecology and Environment of the People’s Republic of China. China mobile source environmental management annual report (2023)[EB/OL]. (2023-12-07)[2024-05-02]. https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml. https://www.mee.gov.cn/hjzl/sthjzk/ydyhjgl/202312/t20231207_1058460.shtml
[2]
Wang J, Rakha H A. Fuel consumption model for heavy duty diesel trucks: model development and testing[J]. Transportation Research Part D: Transport and Environment, 2017, 55: 127-141.
[3]
Chupp W, Zubrow A, Raymond A, et al. MOVES3 sensitivity analysis[R]. Washington D C: Federal Highway Administration, 2023.
[4]
Smit R, Ntziachristos L, Boulter P. Validation of road vehicle and traffic emission models: a review and Meta-analysis[J]. Atmospheric Environment, 2010, 44(25): 2943-2953.
[5]
Fan P, Song G, Zhai Z, et al. Fuel consumption estimation in heavy-duty trucks: integrating vehicle weight into deep-learning frameworks[J]. Transportation Research Part D: Transport and Environment, 2024, 130: 104157.
[6]
阳冬波, 王敏, 邱兆文. 不同公路等级的汽车油耗和排放评估研究[J]. 科学技术与工程, 2017, 17(12): 108-114.
Yang Dongbo, Wang Min, Qiu Zhaowen. Research on evaluation of vehicle fuel consumption and emission of different highway classes[J]. Science Technology and Engineering, 2017, 17(12): 108-114.
[7]
Gavric S, Erdagi I G, Stevanovic A. Environmental assessment of incorrect automated pedestrian detection and common pedestrian timing treatments at signalized intersections[J]. Sustainability, 2024, 16(11): 4487.
[8]
Fan P, Song G, Zhai Z, et al. Fuel consumption estimation in heavy-duty trucks: integrating vehicle weight into deep-learning frameworks[J]. Transportation Research Part D: Transport and Environment, 2024, 130: 104157.
[9]
Ashqer M I, Ashqar H I, Elhenawy M, et al. Evaluating a signalized intersection performance using unmanned aerial Data[J]. Transportation Letters, 2024, 16(5): 452-460.
[10]
Moradi E, Miranda-Moreno L. Vehicular fuel consumption estimation using real-world measures through cascaded machine learning modeling[J]. Transportation Research Part D: Transport and Environment, 2020, 88(11): 102576.
[11]
张雨. 基于LASSO的船舶日油耗预测研究[J]. 中国航海, 2022, 45(4): 129-132, 139.
Zhang Yu. Research on daily fuel consumption prediction of ships based on LASSO[J]. Navigation of China, 2022, 45(4): 129-132, 139.
[12]
Zargarnezhad S, Dashti R, Ahmadi R. Predicting vehicle fuel consumption in energy distribution companies using ANNs[J]. Transportation Research Part D: Transport and Environment, 2019, 74: 174-188.
[13]
Ghasemi M, Qazani M R C, Lennon C W, et al. Analysis and prediction of microbial fuel cell behaviour using MLP and SVR[J]. Journal of the Taiwan Institute of Chemical Engineers, 2023, 151: 105101.
[14]
Wang G, Zhang L, Xu Z, et al. Predictability of vehicle fuel consumption using LSTM: findings from field experiments[J]. Journal of Transportation Engineering, Part A: Systems, 2023, 149(5): 04023030.
[15]
Walnum H J, Simonsen M. Does driving behavior matter? an analysis of fuel consumption data from heavy-duty trucks[J]. Transportation Research Part D, 2015, 36(5): 107-120.
[16]
Hamed M A, Khafagy M H, Badry R M. Fuel consumption prediction model using machine learning[J]. International Journal of Advanced Computer Science and Applications, 2021, 12(11): 406-414.
[17]
Du Y, Wu J, Yang S, et al. Predicting vehicle fuel consumption patterns using floating vehicle data[J]. Journal of Environmental Sciences, 2017, 59: 24-29.
[18]
Xu Z, Wei T, Easa S, et al. Modeling relationship between truck fuel consumption and driving behavior using data from internet of vehicles[J]. Computer-Aided Civil and Infrastructure Engineering, 2018, 33(3): 209-219.
[19]
Wang J, Shen L, Bi Y, et al. Modeling and optimization of a light-duty diesel engine at high altitude with a support vector machine and a genetic algorithm[J]. Fuel, 2021, 285: 119137.
[20]
王一婷, 行本贝, 李彬, . 基于SHARED-LSTM的重型卡车不同加速驾驶行为油耗预测方法[J]. 计算机与现代化, 2023(3): 121-126.
Wang Yiting, Hang Benbei, Li Bin, et al. Fuel consumption prediction method of heavy trucks with different accelerating driving behaviors based on SHARED-LSTM[J]. Computer and Modernization, 2023(3): 121-126.
[21]
王愈轩, 刘尔佳, 黄永章. 基于CNN-LSTM-lightGBM组合的超短期风电功率预测方法[J]. 科学技术与工程, 2022, 22(36): 16067-16074.
Wang Yuxuan, Liu Erjia, Huang Yongzhang. Ultra-short term wind power forecasting method based on CNN-LSTM-lightGBM combination[J]. Science Technology and Engineering, 2019, 22(36): 16067-16074.
[22]
程杰, 陈鼎, 李春, . 基于GWO-CNN-BiLSTM的超短期风电预测[J]. 科学技术与工程, 2023, 23(35): 15091-15099.
Cheng Jie, Chen Ding, Li Chun, et al. Ultra-short term wind power prediction based on GWO-CNN-BiLSTM[J]. Science Technology and Engineering, 2023, 23(35): 15091-15099.
[23]
Li L, Jamieson K, DeSalvo G, et al. Hyperband: a novel bandit-based approach to hyperparameter optimization[J]. Journal of Machine Learning Research, 2018, 18(185): 1-52.
2025年第25卷第9期
PDF下载
268
66
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2403249
  • 接收时间:2024-05-02
  • 首发时间:2025-07-09
  • 出版时间:2025-03-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-02
  • 修回日期:2024-12-23
基金
作者信息
    新疆农业大学交通与物流工程学院, 乌鲁木齐 830052
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403249
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