Article(id=1149773881461923902, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149773869357167407, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2404890, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1719676800000, receivedDateStr=2024-06-30, revisedDate=1738857600000, revisedDateStr=2025-02-07, acceptedDate=null, acceptedDateStr=null, onlineDate=1752057055105, onlineDateStr=2025-07-09, pubDate=1746633600000, pubDateStr=2025-05-08, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752057055105, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752057055105, creator=13701087609, updateTime=1752057055105, updator=13701087609, issue=Issue{id=1149773869357167407, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='13', pageStart='5273', pageEnd='5704', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752057052207, creator=13701087609, updateTime=1768456769392, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1218559268744253990, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149773869357167407, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1218559268744253991, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149773869357167407, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=5535, endPage=5543, ext={EN=ArticleExt(id=1149773881885548611, articleId=1149773881461923902, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=pH Prediction Model of Desulfurization System Based on Bi-directional Gated Recurrent Unit, columnId=1172606851066774020, journalTitle=Science Technology and Engineering, columnName=Papers·Chemical Industry, runingTitle=null, highlight=null, articleAbstract=

Aiming to address the issue of accurately measuring the slurry pH value during the operation of limestone-gypsum wet flue gas desulfurization (WFGD) system, which hinders the efficient operation of WFGD, a prediction model for FGD system pH based on bi-directional gated recirculation unit(BiGRU) has been developed. Firstly, the raw data were cleaned and normalized. Secondly, based on the maximum information coefficient analysis, 13 characteristic values were obtained as input variables and pH as output variables, and a slurry pH model was established. Finally, the model was run and the results were evaluated. Compared to LSTM and GRU, the results indicate that the mean absolute error of this mathematical model decreases by 11.95% and 24.92%, while the root mean square error decreases by 10.64% and 19.49%. Additionally, the coefficient of determination improves by 1.79% and 3.08% respectively. This demonstrates that the BiGRU-based pH predictive model exhibits high accuracy and stability, making it valuable for engineering applications and providing an important reference for predicting pH models in existing desulfurization tower systems.

, correspAuthors=Xian-yong PENG, 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=Jia GAO, Xue-feng TIAN, Xian-yong PENG, Jia-ding JIANG, Min XU), CN=ArticleExt(id=1149773917402915341, articleId=1149773881461923902, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于双向门控循环单元的脱硫系统pH预测模型, columnId=1172606851230351877, journalTitle=科学技术与工程, columnName=论文·化学工业, runingTitle=null, highlight=null, articleAbstract=

针对石灰石-石膏湿法烟气脱硫系统(wet flue gas desulfurization,WFGD)工作过程中浆液pH难以精准测量、不利于WFGD作业的问题,建立一种基于双向门控循环单元的脱硫系统pH预测模型。首先,对原始数据进行清洗和归一化处理;其次,基于最大信息系数分析得出13个特征值为输入变量,pH为输出变量,并建立浆液pH模型;最后,运行模型,并对结果进行评价。研究结果显示,与长短期记忆和门控循环相比,所选用的数学模型的平均绝对误差分别下降了11.95%、24.92%,均方根误差分别下降了10.64%、19.49%,决定系数分别提高了1.79%、3.08%。表明基于双向门控循环单元的pH预测模型具有较高的精确度和稳定性,具有工程应用价值,为现有脱硫塔pH预测模型提供了工程参考。

, correspAuthors=彭献永, authorNote=null, correspAuthorsNote=
* 彭献永(1975—),男,汉族,江苏徐州人,博士,副教授,硕士研究生导师。研究方向:智能发电与综合智慧能源。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=zImCnIs8vNdDP8KrLyf1kw==, magXml=lP/Wox03F26FhwOP2L2ZDg==, pdfUrl=null, pdf=HAXLXgUqnJMcKK9v+eiVjQ==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=px3l4n4M1WBUMZlaa6jVGw==, mapNumber=null, authorCompany=null, fund=null, authors=

高钾(1985—),男,汉族,山西朔州人,博士研究生,讲师。研究方向:燃烧监测与智能发电。E-mail:

, authorsList=高钾, 田雪峰, 彭献永, 蒋甲丁, 徐敏)}, authors=[Author(id=1175114660777964256, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=db23130006p41@cumt.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1175114660891210467, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114660777964256, language=EN, stringName=Jia GAO, firstName=Jia, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China
2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1175114660979290852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114660777964256, language=CN, stringName=高钾, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116
2 新疆工程学院能源工程学院, 乌鲁木齐 830023, bio={"content":"

高钾(1985—),男,汉族,山西朔州人,博士研究生,讲师。研究方向:燃烧监测与智能发电。E-mail:

"}, bioImg=null, bioContent=

高钾(1985—),男,汉族,山西朔州人,博士研究生,讲师。研究方向:燃烧监测与智能发电。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1175114660513723094, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=1, ext=[AuthorCompanyExt(id=1175114660526306007, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China), AuthorCompanyExt(id=1175114660534694616, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116)]), AuthorCompany(id=1175114660597609177, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=2, ext=[AuthorCompanyExt(id=1175114660601803482, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China), AuthorCompanyExt(id=1175114660610192091, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 新疆工程学院能源工程学院, 乌鲁木齐 830023)])]), Author(id=1175114661038011110, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, 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=1175114661113508585, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661038011110, language=EN, stringName=Xue-feng TIAN, firstName=Xue-feng, middleName=null, lastName=TIAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China
2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1175114661197394666, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661038011110, language=CN, stringName=田雪峰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116
2 新疆工程学院能源工程学院, 乌鲁木齐 830023, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1175114660513723094, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=1, ext=[AuthorCompanyExt(id=1175114660526306007, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China), AuthorCompanyExt(id=1175114660534694616, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116)]), AuthorCompany(id=1175114660597609177, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=2, ext=[AuthorCompanyExt(id=1175114660601803482, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China), AuthorCompanyExt(id=1175114660610192091, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 新疆工程学院能源工程学院, 乌鲁木齐 830023)])]), Author(id=1175114661272892140, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=xianyong.peng@cumt.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1175114661356778222, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661272892140, language=EN, stringName=Xian-yong PENG, firstName=Xian-yong, middleName=null, lastName=PENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1175114661444858607, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661272892140, language=CN, stringName=彭献永, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1175114660513723094, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=1, ext=[AuthorCompanyExt(id=1175114660526306007, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China), AuthorCompanyExt(id=1175114660534694616, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116)])]), Author(id=1175114661524550385, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, 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=1175114661604242163, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661524550385, language=EN, stringName=Jia-ding JIANG, firstName=Jia-ding, middleName=null, lastName=JIANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1175114661692322548, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661524550385, language=CN, stringName=蒋甲丁, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 新疆工程学院能源工程学院, 乌鲁木齐 830023, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1175114660597609177, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=2, ext=[AuthorCompanyExt(id=1175114660601803482, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China), AuthorCompanyExt(id=1175114660610192091, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 新疆工程学院能源工程学院, 乌鲁木齐 830023)])]), Author(id=1175114661788791542, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, 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=1175114661868483320, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661788791542, language=EN, stringName=Min XU, firstName=Min, middleName=null, lastName=XU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 China Coal Yudian (Yuhuan) Energy Development Co. , Ltd. , Taizhou 317604, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1175114661943980793, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, authorId=1175114661788791542, language=CN, stringName=徐敏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 中煤玉电(玉环)能源开发有限公司, 台州 317604, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1175114660694078172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=3, ext=[AuthorCompanyExt(id=1175114660698272477, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660694078172, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 China Coal Yudian (Yuhuan) Energy Development Co. , Ltd. , Taizhou 317604, China), AuthorCompanyExt(id=1175114660702466782, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660694078172, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 中煤玉电(玉环)能源开发有限公司, 台州 317604)])])], keywords=[Keyword(id=1175114662057227002, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, orderNo=1, keyword=limestone-gypsum wet flue gas desulfurization system), Keyword(id=1175114662145307387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, orderNo=2, keyword=BiGRU), Keyword(id=1175114662216610556, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, orderNo=3, keyword=prediction model), Keyword(id=1175114662275330813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, orderNo=4, keyword=slurry pH), Keyword(id=1175114662380188414, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, orderNo=1, keyword=石灰石-石膏湿法烟气脱硫系统), Keyword(id=1175114662489240319, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, orderNo=2, keyword=双向门控循环单元神经网络), Keyword(id=1175114662573126400, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, orderNo=3, keyword=预测模型), Keyword(id=1175114662657012481, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, orderNo=4, keyword=浆液pH)], refs=[Reference(id=1175114671397942054, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=27, issue=2, pageStart=27, pageEnd=32, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=柴晋, 乔加飞, 孙灏, journalName=洁净煤技术, refType=null, unstructuredReference=柴晋, 乔加飞, 孙灏, 等. 神经网络算法在脱硫系统优化中的应用进展[J]. 洁净煤技术, 2021, 27(2): 27-32., articleTitle=神经网络算法在脱硫系统优化中的应用进展, refAbstract=null), Reference(id=1175114671490216743, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=27, issue=2, pageStart=27, pageEnd=32, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Chai Jin, Qiao Jiafei, Sun Hao, journalName=Clean Coal Technology, refType=null, unstructuredReference=Chai Jin, Qiao Jiafei, Sun Hao, et al. Progress in the application of neural network algorithm in the optimization of desulfurization system[J]. Clean Coal Technology, 2021, 27(2): 27-32., articleTitle=Progress in the application of neural network algorithm in the optimization of desulfurization system, refAbstract=null), Reference(id=1175114671565714216, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2019, volume=4, issue=11, pageStart=929, pageEnd=938, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Tang L, Qu J B, Mi Z F, journalName=Nature Energy, refType=null, unstructuredReference=Tang L, Qu J B, Mi Z F, et al. Substantial emission reductions from Chinese power plants after the introduction of ultra-low emissions standards[J]. Nature Energy, 2019, 4(11): 929-938., articleTitle=Substantial emission reductions from Chinese power plants after the introduction of ultra-low emissions standards, refAbstract=null), Reference(id=1175114671725097769, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=29, issue=S2, pageStart=406, pageEnd=414, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=孔若琪, 崔琳, 董勇, journalName=洁净煤技术, refType=null, unstructuredReference=孔若琪, 崔琳, 董勇. 机器学习算法在脱硫系统智能运行及优化中的应用[J]. 洁净煤技术, 2023, 29(S2): 406-414., articleTitle=机器学习算法在脱硫系统智能运行及优化中的应用, refAbstract=null), Reference(id=1175114671922230058, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=29, issue=S2, pageStart=406, pageEnd=414, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Kong Ruoqi, Cui Lin, Dong Yong, journalName=Clean Coal Technology, refType=null, unstructuredReference=Kong Ruoqi, Cui Lin, Dong Yong. Application of machine learning algorithm in intelligent operation and optimization of desulfurization system[J]. Clean Coal Technology, 2023, 29(S2): 406-414., articleTitle=Application of machine learning algorithm in intelligent operation and optimization of desulfurization system, refAbstract=null), Reference(id=1175114672018699051, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2019, volume=39, issue=10, pageStart=834, pageEnd=839, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=康英伟, 孙智滨, 常俊, journalName=动力工程学报, refType=null, unstructuredReference=康英伟, 孙智滨, 常俊. 基于改进差分进化算法的湿法烟气脱硫系统模型辨识[J]. 动力工程学报, 2019, 39(10): 834-839., articleTitle=基于改进差分进化算法的湿法烟气脱硫系统模型辨识, refAbstract=null), Reference(id=1175114672098390828, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2019, volume=39, issue=10, pageStart=834, pageEnd=839, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Kang Yingwei, Sun Zhibin, Chang Jun, journalName=Journal of Power Engineering, refType=null, unstructuredReference=Kang Yingwei, Sun Zhibin, Chang Jun. Model identification of wet flue gas desulfurization system based on improved differential evolutionary algorithm[J]. Journal of Power Engineering, 2019, 39(10): 834-839., articleTitle=Model identification of wet flue gas desulfurization system based on improved differential evolutionary algorithm, refAbstract=null), Reference(id=1175114672186471213, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2017, volume=37, issue=11, pageStart=925, pageEnd=930, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=陈浩, 骆仲泱, 江建平, journalName=动力工程学报, refType=null, unstructuredReference=陈浩, 骆仲泱, 江建平, 等. 电厂湿法烟气脱硫颗粒物排放特性的实验研究[J]. 动力工程学报, 2017, 37(11): 925-930, 937., articleTitle=电厂湿法烟气脱硫颗粒物排放特性的实验研究, refAbstract=null), Reference(id=1175114672282940206, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2017, volume=37, issue=11, pageStart=925, pageEnd=930, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Chen Hao, Luo Zhongyang, Jiang Jianping, journalName=Journal of Power Engineering, refType=null, unstructuredReference=Chen Hao, Luo Zhongyang, Jiang Jianping, et al. Experimental study on the emission characteristics of particulate matter from wet flue gas desulfurization in a power plant[J]. Journal of Power Engineering, 2017, 37(11): 925-930, 937., articleTitle=Experimental study on the emission characteristics of particulate matter from wet flue gas desulfurization in a power plant, refAbstract=null), Reference(id=1175114672375214895, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=94, issue=null, pageStart=167, pageEnd=175, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=Liu G F, Zhang Y, Shen D K, journalName=Journal of the Energy Institute, refType=null, unstructuredReference=Liu G F, Zhang Y, Shen D K, et al. Anticipatory NH3 injection control for SCR system based on the prediction of the inlet NOx concentration[J]. Journal of the Energy Institute, 2021, 94: 167-175., articleTitle=Anticipatory NH3 injection control for SCR system based on the prediction of the inlet NOx concentration, refAbstract=null), Reference(id=1175114672459100976, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=37, issue=8, pageStart=60, pageEnd=66, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=李建强, 杨红, 牛成林, journalName=电力科学与工程, refType=null, unstructuredReference=李建强, 杨红, 牛成林. 基于改进即时学习算法的吸收塔pH软测量技术研究[J]. 电力科学与工程, 2021, 37(8): 60-66., articleTitle=基于改进即时学习算法的吸收塔pH软测量技术研究, refAbstract=null), Reference(id=1175114672534598449, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=37, issue=8, pageStart=60, pageEnd=66, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=Li Jianqiang, Yang Hong, Niu Chenglin, journalName=Electric Power Science and Engineering, refType=null, unstructuredReference=Li Jianqiang, Yang Hong, Niu Chenglin. Research on soft measurement technology of pH value in absorption tower based on improved instant learning algorithm[J]. Electric Power Science and Engineering, 2021, 37(8): 60-66., articleTitle=Research on soft measurement technology of pH value in absorption tower based on improved instant learning algorithm, refAbstract=null), Reference(id=1175114672631067442, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=40, issue=3, pageStart=1689, pageEnd=1698, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=马双忱, 林宸雨, 周权, journalName=化工进展, refType=null, unstructuredReference=马双忱, 林宸雨, 周权. 基于深度神经网络的脱硫系统预测模型及应用[J]. 化工进展, 2021, 40(3): 1689-1698., articleTitle=基于深度神经网络的脱硫系统预测模型及应用, refAbstract=null), Reference(id=1175114672803033907, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=40, issue=3, pageStart=1689, pageEnd=1698, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=Ma Shuangchen, Lin Chenyu, Zhou Quan, journalName=Advances in Chemical Engineering, refType=null, unstructuredReference=Ma Shuangchen, Lin Chenyu, Zhou Quan. Prediction model and application of desulfurization system based on deep neural network[J]. Advances in Chemical Engineering, 2021, 40(3): 1689-1698., articleTitle=Prediction model and application of desulfurization system based on deep neural network, refAbstract=null), Reference(id=1175114672937251636, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2020, volume=39, issue=8, pageStart=62, pageEnd=66, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=金秀章, 景昊, journalName=信息技术与网络安全, refType=null, unstructuredReference=金秀章, 景昊. 基于 LSTM 的湿法烟气脱硫浆液pH建模[J]. 信息技术与网络安全, 2020, 39(8): 62-66., articleTitle=基于 LSTM 的湿法烟气脱硫浆液pH建模, refAbstract=null), Reference(id=1175114673058886453, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2020, volume=39, issue=8, pageStart=62, pageEnd=66, url=null, language=null, rfNumber=[9], rfOrder=15, authorNames=Jin Xiuzhang, Jing Hao, journalName=Information Technology and Network Security, refType=null, unstructuredReference=Jin Xiuzhang, Jing Hao. LSTM-based pH modeling of wet flue gas desulfurization slurry[J]. Information Technology and Network Security, 2020, 39(8): 62-66., articleTitle=LSTM-based pH modeling of wet flue gas desulfurization slurry, refAbstract=null), Reference(id=1175114673184715574, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2022, volume=207, issue=null, pageStart=117986, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=Gupta U, Bhattacharjee V, Bishnu S, journalName=Expert Systems with Applications, refType=null, unstructuredReference=Gupta U, Bhattacharjee V, Bishnu S. Stock Net-GRU based stock index prediction[J]. Expert Systems with Applications, 2022, 207: 117986., articleTitle=Stock Net-GRU based stock index prediction, refAbstract=null), Reference(id=1175114673314738999, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=18, pageStart=7824, pageEnd=7830, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=郝晴, 迟涛, 于政军, journalName=科学技术与工程, refType=null, unstructuredReference=郝晴, 迟涛, 于政军, 等. 基于门控循环单元神经网络的浆液pH的预测建模[J]. 科学技术与工程, 2023, 23(18): 7824-7830., articleTitle=基于门控循环单元神经网络的浆液pH的预测建模, refAbstract=null), Reference(id=1175114673377653560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=18, pageStart=7824, pageEnd=7830, url=null, language=null, rfNumber=[11], rfOrder=18, authorNames=Hao Qing, Chi Tao, Yu Zhengjun, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Hao Qing, Chi Tao, Yu Zhengjun, et al. Predictive modeling of slurry pH based on gated recurrent unit neural network[J]. Science Technology and Engineering, 2023, 23(18): 7824-7830., articleTitle=Predictive modeling of slurry pH based on gated recurrent unit neural network, refAbstract=null), Reference(id=1175114673478316857, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2019, volume=40, issue=1, pageStart=51, pageEnd=55, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=Li W D, Gao H H, Cai W F, journalName=Power Generation Technology, refType=null, unstructuredReference=Li W D, Gao H H, Cai W F. Scaling analysis and preventive measures of limestone-gypsum wet flue gas desulfurization[J]. Power Generation Technology, 2019, 40(1): 51-55., articleTitle=Scaling analysis and preventive measures of limestone-gypsum wet flue gas desulfurization, refAbstract=null), Reference(id=1175114673587368762, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2020, volume=39, issue=S1, pageStart=242, pageEnd=249, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=曹建宗, 刘琦, 陈文通, journalName=化工进展, refType=null, unstructuredReference=曹建宗, 刘琦, 陈文通, 等. 典型湿法脱硫系统存在的问题及人工智能在优化运行中的应用[J]. 化工进展, 2020, 39(S1): 242-249., articleTitle=典型湿法脱硫系统存在的问题及人工智能在优化运行中的应用, refAbstract=null), Reference(id=1175114673654477627, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2020, volume=39, issue=S1, pageStart=242, pageEnd=249, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=Cao Jianzong, Liu Qi, Chen Wentong, journalName=Progress in Chemical Engineering, refType=null, unstructuredReference=Cao Jianzong, Liu Qi, Chen Wentong, et al. Problems of typical wet desulfurization systems and the application of artificial intelligence in optimized operation[J]. Progress in Chemical Engineering, 2020, 39(S1): 242-249., articleTitle=Problems of typical wet desulfurization systems and the application of artificial intelligence in optimized operation, refAbstract=null), Reference(id=1175114673746752316, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=22, authorNames=董家伟, journalName=湿法烟气脱硫pH的模型预测控制, refType=null, unstructuredReference=董家伟. 湿法烟气脱硫pH的模型预测控制[D]. 北京: 北京工业大学, 2022., articleTitle=null, refAbstract=null), Reference(id=1175114673830638397, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=Dong Jiawei, journalName=Model predictive control of pH in wet flue gas desulfurization, refType=null, unstructuredReference=Dong Jiawei. Model predictive control of pH in wet flue gas desulfurization[D]. Beijing: Beijing Institute of Technology, 2022., articleTitle=null, refAbstract=null), Reference(id=1175114673910330174, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=24, authorNames=Cho K, Van Merriënboer B, Gulcehre C, journalName=ArXiv Preprint ArXiv, refType=null, unstructuredReference=Cho K, Van Merriënboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder forstatistical machine translation[J]. ArXiv Preprint ArXiv, 2014: 1406. 1078., articleTitle=Learning phrase representations using RNN encoder-decoder forstatistical machine translation, refAbstract=null), Reference(id=1175114674023576383, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=685, pageEnd=690, url=null, language=null, rfNumber=[16], rfOrder=25, authorNames=Wang D, Mao K, journalName=2018 IEEE Third International Conference on Data Science in Cyberspace (DSC), refType=null, unstructuredReference=Wang D, Mao K. Multimodal object classification using bidirectional gated recurrent unit networks[C]// 2018 IEEE Third International Conference on Data Science in Cyberspace (DSC). Guangzhou: IEEE, 2018: 685-690., articleTitle=Multimodal object classification using bidirectional gated recurrent unit networks, refAbstract=null), Reference(id=1175114674120045376, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=12, pageStart=4370, pageEnd=4376, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=赵兵, 王增平, 纪维佳, journalName=电网技术, refType=null, unstructuredReference=赵兵, 王增平, 纪维佳, 等. 基于注意力机制的CNN-GRU短期电力负荷预测方法[J]. 电网技术, 2019, 43(12): 4370-4376., articleTitle=基于注意力机制的CNN-GRU短期电力负荷预测方法, refAbstract=null), Reference(id=1175114674216514369, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=12, pageStart=4370, pageEnd=4376, url=null, language=null, rfNumber=[17], rfOrder=27, authorNames=Zhao Bing, Wang Zengping, Ji Weijia, journalName=Grid Technology, refType=null, unstructuredReference=Zhao Bing, Wang Zengping, Ji Weijia, et al. CNN-GRU short-term power load forecasting method based on attention mechanism[J]. Grid Technology, 2019, 43(12): 4370-4376., articleTitle=CNN-GRU short-term power load forecasting method based on attention mechanism, refAbstract=null), Reference(id=1175114674317177666, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=50, issue=12, pageStart=3519, pageEnd=3523, url=null, language=null, rfNumber=[18], rfOrder=28, authorNames=蒋星明, 曹顺安, 王民军, journalName=应用化工, refType=null, unstructuredReference=蒋星明, 曹顺安, 王民军, 等. 双向门控循环神经网络的SO2排放浓度预测模型[J]. 应用化工, 2021, 50(12): 3519-3523., articleTitle=双向门控循环神经网络的SO2排放浓度预测模型, refAbstract=null), Reference(id=1175114674413646659, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=50, issue=12, pageStart=3519, pageEnd=3523, url=null, language=null, rfNumber=[18], rfOrder=29, authorNames=Jiang Xingming, Cao Shun'an, Wang Minjun, journalName=Applied Chemical Engineering, refType=null, unstructuredReference=Jiang Xingming, Cao Shun'an, Wang Minjun, et al. SO2 emission concentration prediction modeling by two-way gated recurrent neural network[J]. Applied Chemical Engineering, 2021, 50(12): 3519-3523., articleTitle=SO2 emission concentration prediction modeling by two-way gated recurrent neural network, refAbstract=null), Reference(id=1175114674501727044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=30, authorNames=Chung J, Gulcehre C, Cho K H, journalName=ArXiv Preprint ArXiv, refType=null, unstructuredReference=Chung J, Gulcehre C, Cho K H, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J]. ArXiv Preprint ArXiv, 2014: 1412. 3555., articleTitle=Empirical evaluation of gated recurrent neural networks on sequence modeling, refAbstract=null), Reference(id=1175114674581418821, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=31, authorNames=许崇轩, journalName=基于相空间重构和双向门控循环单元神经网络的海水指标预测研究, refType=null, unstructuredReference=许崇轩. 基于相空间重构和双向门控循环单元神经网络的海水指标预测研究[D]. 秦皇岛: 燕山大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1175114674661110598, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=32, authorNames=Xu Chongxuan, journalName=Research on seawater index prediction based on phase space reconstruction and two-way gated recurrent unit neural network, refType=null, unstructuredReference=Xu Chongxuan. Research on seawater index prediction based on phase space reconstruction and two-way gated recurrent unit neural network[D]. Qinhuangdao: Yanshan University, 2023., articleTitle=null, refAbstract=null), Reference(id=1175114674749190983, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=3, pageStart=13, pageEnd=22, url=null, language=null, rfNumber=[21], rfOrder=33, authorNames=杨寒雨, 赵晓永, 王磊, journalName=计算机工程与应用, refType=null, unstructuredReference=杨寒雨, 赵晓永, 王磊. 数据归一化方法综述[J]. 计算机工程与应用, 2023, 59(3): 13-22., articleTitle=数据归一化方法综述, refAbstract=null), Reference(id=1175114674845659976, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=3, pageStart=13, pageEnd=22, url=null, language=null, rfNumber=[21], rfOrder=34, authorNames=Yang Hanyu, Zhao Xiaoyong, Wang Lei, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=Yang Hanyu, Zhao Xiaoyong, Wang Lei. A review of data normalization methods[J]. Computer Engineering and Applications, 2023, 59(3): 13-22., articleTitle=A review of data normalization methods, refAbstract=null), Reference(id=1175114674925351753, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=13, pageStart=66, pageEnd=72, url=null, language=null, rfNumber=[22], rfOrder=35, authorNames=孙坤, 尹晓红, journalName=电子测量技术, refType=null, unstructuredReference=孙坤, 尹晓红. 基于数据去噪和CNN-BiGRU的SO2排放预测[J]. 电子测量技术, 2023, 46(13): 66-72., articleTitle=基于数据去噪和CNN-BiGRU的SO2排放预测, refAbstract=null), Reference(id=1175114675000849226, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=13, pageStart=66, pageEnd=72, url=null, language=null, rfNumber=[22], rfOrder=36, authorNames=Sun Kun, Yin Xiaohong, journalName=Electronic Measurement Technology, refType=null, unstructuredReference=Sun Kun, Yin Xiaohong. SO2 emission prediction based on data denoising and CNN-BiGRU[J]. Electronic Measurement Technology, 2023, 46(13): 66-72., articleTitle=SO2 emission prediction based on data denoising and CNN-BiGRU, refAbstract=null), Reference(id=1175114675185398603, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=72, issue=3, pageStart=1480, pageEnd=1486, url=null, language=null, rfNumber=[23], rfOrder=37, authorNames=顾俊发, 许明阳, 马方圆, journalName=化工学报, refType=null, unstructuredReference=顾俊发, 许明阳, 马方圆, 等. 基于MIC的支持向量回归及其在化工过程中的应用[J]. 化工学报, 2021, 72(3): 1480-1486., articleTitle=基于MIC的支持向量回归及其在化工过程中的应用, refAbstract=null), Reference(id=1175114675265090380, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=72, issue=3, pageStart=1480, pageEnd=1486, url=null, language=null, rfNumber=[23], rfOrder=38, authorNames=Gu Junfa, Xu Mingyang, Ma Fangyuan, journalName=Journal of Chemical Engineering, refType=null, unstructuredReference=Gu Junfa, Xu Mingyang, Ma Fangyuan, et al. MIC-based support vector regression and its application to chemical processes[J]. Journal of Chemical Engineering, 2021, 72(3): 1480-1486., articleTitle=MIC-based support vector regression and its application to chemical processes, refAbstract=null), Reference(id=1175114675319616333, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2011, volume=334, issue=6062, pageStart=1518, pageEnd=1524, url=null, language=null, rfNumber=[24], rfOrder=39, authorNames=David N, Reshef Y A, Finucane H K, journalName=Science, refType=null, unstructuredReference=David N, Reshef Y A, Finucane H K, et al. Detecting novel associntionsin large data sets[J]. Science, 2011, 334(6062): 1518-1524., articleTitle=Detecting novel associntionsin large data sets, refAbstract=null), Reference(id=1175114675416085326, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=29, pageStart=12616, pageEnd=12622, url=null, language=null, rfNumber=[25], rfOrder=40, authorNames=吴磊, 康英伟, journalName=科学技术与工程, refType=null, unstructuredReference=吴磊, 康英伟. 基于长短时记忆-反向传播级联模型的湿法脱硫系统性能预测[J]. 科学技术与工程, 2021, 21(29): 12616-12622., articleTitle=基于长短时记忆-反向传播级联模型的湿法脱硫系统性能预测, refAbstract=null), Reference(id=1175114675478999887, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=29, pageStart=12616, pageEnd=12622, url=null, language=null, rfNumber=[25], rfOrder=41, authorNames=Wu Lei, Kang Yingwei, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wu Lei, Kang Yingwei. Performance prediction of wet desulfurization system based on long-short-term memory-back propagation cascade model[J]. Science Technology and Engineering, 2021, 21(29): 12616-12622., articleTitle=Performance prediction of wet desulfurization system based on long-short-term memory-back propagation cascade model, refAbstract=null)], funds=[Fund(id=1175114671095952164, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, awardId=72361033, language=CN, fundingSource=国家自然科学基金(72361033), fundOrder=null, country=null), Fund(id=1175114671205004069, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, awardId=HNKJ23-HF31, language=CN, fundingSource=中国华能集团公司总部科技资助项目(HNKJ23-HF31), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1175114660513723094, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=1, ext=[AuthorCompanyExt(id=1175114660526306007, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China), AuthorCompanyExt(id=1175114660534694616, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660513723094, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国矿业大学低碳能源与动力工程学院, 徐州 221116)]), AuthorCompany(id=1175114660597609177, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=2, ext=[AuthorCompanyExt(id=1175114660601803482, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China), AuthorCompanyExt(id=1175114660610192091, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660597609177, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 新疆工程学院能源工程学院, 乌鲁木齐 830023)]), AuthorCompany(id=1175114660694078172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, xref=3, ext=[AuthorCompanyExt(id=1175114660698272477, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660694078172, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 China Coal Yudian (Yuhuan) Energy Development Co. , Ltd. , Taizhou 317604, China), AuthorCompanyExt(id=1175114660702466782, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, companyId=1175114660694078172, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 中煤玉电(玉环)能源开发有限公司, 台州 317604)])], figs=[ArticleFig(id=1175114662862533378, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.1, caption=Process flow diagram of WFGD, figureFileSmall=e0D/D7+oFg6NYxxIwTc3qA==, figureFileBig=zQFdWVW+/8W3dOkri+0v1g==, tableContent=null), ArticleFig(id=1175114662942225155, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图1, caption=石灰石-石膏湿法脱硫工艺流程图, figureFileSmall=e0D/D7+oFg6NYxxIwTc3qA==, figureFileBig=zQFdWVW+/8W3dOkri+0v1g==, tableContent=null), ArticleFig(id=1175114663017722628, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.2, caption=GRU structure diagram, figureFileSmall=wUyKBw0wJr2QqcMAOuIQNQ==, figureFileBig=xdbzxarL0jjSeRilIr+tjQ==, tableContent=null), ArticleFig(id=1175114663076442885, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图2, caption=GRU结构图, figureFileSmall=wUyKBw0wJr2QqcMAOuIQNQ==, figureFileBig=xdbzxarL0jjSeRilIr+tjQ==, tableContent=null), ArticleFig(id=1175114663143551750, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.3, caption=BiGRU structure diagram, figureFileSmall=seb2tW2wBQ30QjPwAyGiWg==, figureFileBig=krHnUNrzb3bc+/3DfRBKYw==, tableContent=null), ArticleFig(id=1175114663240020743, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图3, caption=BiGRU结构图, figureFileSmall=seb2tW2wBQ30QjPwAyGiWg==, figureFileBig=krHnUNrzb3bc+/3DfRBKYw==, tableContent=null), ArticleFig(id=1175114663315518216, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.4, caption=Sampling chart of raw data of unit load, figureFileSmall=4pxnoqiiZ/tQMmwIa+8/+g==, figureFileBig=ByXg53eHcBTOAtd8hE/1Og==, tableContent=null), ArticleFig(id=1175114663382627081, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图4, caption=机组负荷原始数据采样图, figureFileSmall=4pxnoqiiZ/tQMmwIa+8/+g==, figureFileBig=ByXg53eHcBTOAtd8hE/1Og==, tableContent=null), ArticleFig(id=1175114665886626570, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.5, caption=Sampling chart of raw slurry pH data, figureFileSmall=fThuzfapuzzKXoU6HgIHZg==, figureFileBig=KhnPuk7MYwgT8as/aAgZtQ==, tableContent=null), ArticleFig(id=1175114665999872779, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图5, caption=浆液pH原始数据采样图, figureFileSmall=fThuzfapuzzKXoU6HgIHZg==, figureFileBig=KhnPuk7MYwgT8as/aAgZtQ==, tableContent=null), ArticleFig(id=1175114666075370252, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.6, caption=Eigenvalue thermogram of desulfurization system based on MIC, figureFileSmall=32w4lrqQVM3zpf0isFYMxw==, figureFileBig=gmRsuuFXNLixyEuswMUe5g==, tableContent=null), ArticleFig(id=1175114666146673421, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图6, caption=基于MIC的脱硫系统特征值热力图, figureFileSmall=32w4lrqQVM3zpf0isFYMxw==, figureFileBig=gmRsuuFXNLixyEuswMUe5g==, tableContent=null), ArticleFig(id=1175114666276696846, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.7, caption=Structure of BiGRU based pH prediction model, figureFileSmall=ua/GCC2G9Z49pT7AB61pog==, figureFileBig=Z7/luryN69Ar2voAeYfQJQ==, tableContent=null), ArticleFig(id=1175114666364777231, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图7, caption=基于BiGRU的pH预测模型结构图, figureFileSmall=ua/GCC2G9Z49pT7AB61pog==, figureFileBig=Z7/luryN69Ar2voAeYfQJQ==, tableContent=null), ArticleFig(id=1175114666461246224, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.8, caption=Prediction results before and after screening of input feature variables, figureFileSmall=yLtl9BvtsXibV3Fa2xvoJg==, figureFileBig=PEa0zDqOO1D128yG9WsYfg==, tableContent=null), ArticleFig(id=1175114666545132305, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图8, caption=输入特征变量筛选前后的预测结果, figureFileSmall=yLtl9BvtsXibV3Fa2xvoJg==, figureFileBig=PEa0zDqOO1D128yG9WsYfg==, tableContent=null), ArticleFig(id=1175114666704515858, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.9, caption=Prediction error before and after screening of input feature variables, figureFileSmall=Nl8A8jpfodavFKIoRfK3Kg==, figureFileBig=Yr1VKrZ9xDK7EmGv0vu9Ng==, tableContent=null), ArticleFig(id=1175114667010700051, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图9, caption=输入特征变量筛选前后的预测误差, figureFileSmall=Nl8A8jpfodavFKIoRfK3Kg==, figureFileBig=Yr1VKrZ9xDK7EmGv0vu9Ng==, tableContent=null), ArticleFig(id=1175114667115557652, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.10, caption=Predicted and true values for each model, figureFileSmall=vVMi7lfe/7p6bICJ7t70YA==, figureFileBig=8LQ+WPOtvYnx4SsCpP7CiQ==, tableContent=null), ArticleFig(id=1175114667182666517, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图10, caption=各模型预测值与真实值, figureFileSmall=vVMi7lfe/7p6bICJ7t70YA==, figureFileBig=8LQ+WPOtvYnx4SsCpP7CiQ==, tableContent=null), ArticleFig(id=1175114667287524118, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Fig.11, caption=Prediction error for each model, figureFileSmall=vpw7jk1ybPbXtjar4uW1qQ==, figureFileBig=QskOHAREQxGhaSx6CaVrWA==, tableContent=null), ArticleFig(id=1175114667396576023, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=图11, caption=各模型预测误差, figureFileSmall=vpw7jk1ybPbXtjar4uW1qQ==, figureFileBig=QskOHAREQxGhaSx6CaVrWA==, tableContent=null), ArticleFig(id=1175114667509822232, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Table 1, caption=

Ranking of eigenvalues with correlation coefficients greater than 60% with pH

, figureFileSmall=null, figureFileBig=null, tableContent=
变量 相关性 排序
再循环泵5电流 0.873 337 302 1
实际脱硫效率 0.865 925 829 2
吸收塔给浆密度 0.849 549 702 3
出口粉尘含量 0.829 967 304 4
进口烟气温度 0.809 839 438 5
石灰石耗量 0.788 799 965 6
机组负荷 0.779 313 141 7
吸收塔给浆体积流量 0.778 568 806 8
锅炉总风量 0.744 143 442 9
进口烟气含氧量 0.731 875 023 10
FGD出口氧量 0.726 883 095 11
氧化风机流量 0.725 112 664 12
进口烟气SO2浓度 0.677 967 568 13
), ArticleFig(id=1175114667581125401, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=表1, caption=

与pH的相关系数大于60%的特征值排序表

, figureFileSmall=null, figureFileBig=null, tableContent=
变量 相关性 排序
再循环泵5电流 0.873 337 302 1
实际脱硫效率 0.865 925 829 2
吸收塔给浆密度 0.849 549 702 3
出口粉尘含量 0.829 967 304 4
进口烟气温度 0.809 839 438 5
石灰石耗量 0.788 799 965 6
机组负荷 0.779 313 141 7
吸收塔给浆体积流量 0.778 568 806 8
锅炉总风量 0.744 143 442 9
进口烟气含氧量 0.731 875 023 10
FGD出口氧量 0.726 883 095 11
氧化风机流量 0.725 112 664 12
进口烟气SO2浓度 0.677 967 568 13
), ArticleFig(id=1175114667673400090, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Table 2, caption=

Model errors correspond to different MiniBatchSize

, figureFileSmall=null, figureFileBig=null, tableContent=
MiniBatchSize 训练集 测试集
MAE RMSE MAE RMSE
64 0.032 377 0.042 151 0.042 826 0.047 467
128 0.029 997 0.039 027 0.017 399 0.022 880
256 0.023 443 0.027 724 0.013 904 0.018 311
512 0.030 755 0.040 974 0.019 725 0.023 707
), ArticleFig(id=1175114667744703259, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=表2, caption=

不同 MiniBatchSize 对应模型误差

, figureFileSmall=null, figureFileBig=null, tableContent=
MiniBatchSize 训练集 测试集
MAE RMSE MAE RMSE
64 0.032 377 0.042 151 0.042 826 0.047 467
128 0.029 997 0.039 027 0.017 399 0.022 880
256 0.023 443 0.027 724 0.013 904 0.018 311
512 0.030 755 0.040 974 0.019 725 0.023 707
), ArticleFig(id=1175114667866338076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Table 3, caption=

Error of the corresponding models for the different MaxEpochs

, figureFileSmall=null, figureFileBig=null, tableContent=
MaxEpochs 训练集 测试集
MAE RMSE MAE RMSE
100 0.035 586 0.046 405 0.032 147 0.039 754
150 0.032 079 0.044 118 0.039 000 0.042 254
200 0.025 443 0.033 724 0.015 904 0.019 311
250 0.029 465 0.038 428 0.016 930 0.019 613
300 0.031 442 0.040 481 0.015 629 0.019 781
), ArticleFig(id=1175114667962807069, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=表3, caption=

不同 MaxEpochs 对应模型的误差

, figureFileSmall=null, figureFileBig=null, tableContent=
MaxEpochs 训练集 测试集
MAE RMSE MAE RMSE
100 0.035 586 0.046 405 0.032 147 0.039 754
150 0.032 079 0.044 118 0.039 000 0.042 254
200 0.025 443 0.033 724 0.015 904 0.019 311
250 0.029 465 0.038 428 0.016 930 0.019 613
300 0.031 442 0.040 481 0.015 629 0.019 781
), ArticleFig(id=1175114668046693150, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Table 4, caption=

Error of the corresponding models for the different InitialLearnRate

, figureFileSmall=null, figureFileBig=null, tableContent=
InitialLearnRate 训练集 测试集
MAE RMSE MAE RMSE
0.1 0.066 508 0.074 725 0.021 249 0.068 303
0.05 0.036 522 0.046 520 0.020 790 0.060 191
0.01 0.029 830 0.039 101 0.016 772 0.020 949
0.005 0.031 196 0.040 682 0.018 476 0.023 638
0.001 0.038 914 0.051 834 0.050 660 0.056 773
), ArticleFig(id=1175114670496166687, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=表4, caption=

不同 InitialLearnRate 对应模型的误差

, figureFileSmall=null, figureFileBig=null, tableContent=
InitialLearnRate 训练集 测试集
MAE RMSE MAE RMSE
0.1 0.066 508 0.074 725 0.021 249 0.068 303
0.05 0.036 522 0.046 520 0.020 790 0.060 191
0.01 0.029 830 0.039 101 0.016 772 0.020 949
0.005 0.031 196 0.040 682 0.018 476 0.023 638
0.001 0.038 914 0.051 834 0.050 660 0.056 773
), ArticleFig(id=1175114670642967328, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Table 5, caption=

Evaluation indicator values and running time of prediction results before and after screening

, figureFileSmall=null, figureFileBig=null, tableContent=
特征值 MAE RMSE R2 用时/s
筛选前 0.017 39 0.021 72 0.971 17 151
筛选后 0.014 959 0.018 721 0.987 71 133
), ArticleFig(id=1175114670739436321, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=表5, caption=

筛选前后预测结果评价指标值和运行时间

, figureFileSmall=null, figureFileBig=null, tableContent=
特征值 MAE RMSE R2 用时/s
筛选前 0.017 39 0.021 72 0.971 17 151
筛选后 0.014 959 0.018 721 0.987 71 133
), ArticleFig(id=1175114670827516706, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=EN, label=Table 6, caption=

Comparison of prediction results of the models

, figureFileSmall=null, figureFileBig=null, tableContent=
样本编号 pH真实值 BiGRU LSTM GRU
pH预测值 预测误差 pH预测值 预测误差 pH预测值 预测误差
100 5.147 461 5.166 43 0.018 969 5.184 723 9 0.037 262 9 5.189 935 2 0.042 474 2
400 5.118 621 5.097 01 -0.021 611 5.093 468 2 -0.025 152 8 5.072 410 6 -0.046 210 4
700 5.349 335 5.329 58 -0.019 755 5.319 975 9 -0.029 359 1 5.314 964 5 -0.034 370 5
1 000 5.291 656 5.294 37 0.002 714 5.281 096 -0.010 56 5.265 517 2 -0.026 138 8
1 300 5.291 656 5.287 644 4 -0.004 011 6 5.260 425 6 -0.031 230 4 5.260 26 -0.031 396
1 600 5.538 391 5.538 793 1 0.000 402 1 5.527 33 -0.011 061 5.531 103 6 -0.007 287 4
1 900 5.624 908 5.621 95 -0.002 958 5.636 314 4 0.011 406 4 5.612 349 5 -0.012 558 5
MAE 0.013 904 0.015 790 0.018 520
RMSE 0.018 311 0.020 491 0.022 743
R2 0.988 71 0.971 29 0.959 15
), ArticleFig(id=1175114670944957219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149773881461923902, language=CN, label=表6, caption=

各模型的预测结果比较

, figureFileSmall=null, figureFileBig=null, tableContent=
样本编号 pH真实值 BiGRU LSTM GRU
pH预测值 预测误差 pH预测值 预测误差 pH预测值 预测误差
100 5.147 461 5.166 43 0.018 969 5.184 723 9 0.037 262 9 5.189 935 2 0.042 474 2
400 5.118 621 5.097 01 -0.021 611 5.093 468 2 -0.025 152 8 5.072 410 6 -0.046 210 4
700 5.349 335 5.329 58 -0.019 755 5.319 975 9 -0.029 359 1 5.314 964 5 -0.034 370 5
1 000 5.291 656 5.294 37 0.002 714 5.281 096 -0.010 56 5.265 517 2 -0.026 138 8
1 300 5.291 656 5.287 644 4 -0.004 011 6 5.260 425 6 -0.031 230 4 5.260 26 -0.031 396
1 600 5.538 391 5.538 793 1 0.000 402 1 5.527 33 -0.011 061 5.531 103 6 -0.007 287 4
1 900 5.624 908 5.621 95 -0.002 958 5.636 314 4 0.011 406 4 5.612 349 5 -0.012 558 5
MAE 0.013 904 0.015 790 0.018 520
RMSE 0.018 311 0.020 491 0.022 743
R2 0.988 71 0.971 29 0.959 15
)], 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.1671-1815.2404890, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2404890, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2404890, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2404890, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于双向门控循环单元的脱硫系统pH预测模型
收藏切换
PDF下载
高钾 1, 2 , 田雪峰 1, 2 , 彭献永 1, * , 蒋甲丁 2 , 徐敏 3
科学技术与工程 | 论文·化学工业 2025,25(13): 5535-5543
收起
收藏切换
科学技术与工程 | 论文·化学工业 2025, 25(13): 5535-5543
基于双向门控循环单元的脱硫系统pH预测模型
全屏
高钾1, 2 , 田雪峰1, 2, 彭献永1, * , 蒋甲丁2, 徐敏3
作者信息
  • 1 中国矿业大学低碳能源与动力工程学院, 徐州 221116
  • 2 新疆工程学院能源工程学院, 乌鲁木齐 830023
  • 3 中煤玉电(玉环)能源开发有限公司, 台州 317604
  • 高钾(1985—),男,汉族,山西朔州人,博士研究生,讲师。研究方向:燃烧监测与智能发电。E-mail:

通讯作者:

* 彭献永(1975—),男,汉族,江苏徐州人,博士,副教授,硕士研究生导师。研究方向:智能发电与综合智慧能源。E-mail:
pH Prediction Model of Desulfurization System Based on Bi-directional Gated Recurrent Unit
Jia GAO1, 2 , Xue-feng TIAN1, 2, Xian-yong PENG1, * , Jia-ding JIANG2, Min XU3
Affiliations
  • 1 School of Low-carbon Energy and Power Engineering, China University of Mining and Technology, Xuzhou 221116, China
  • 2 School of Energy Engineering, Xinjiang Institute of Engineering, Urumqi 830023, China
  • 3 China Coal Yudian (Yuhuan) Energy Development Co. , Ltd. , Taizhou 317604, China
出版时间: 2025-05-08 doi: 10.12404/j.issn.1671-1815.2404890
文章导航
收藏切换

针对石灰石-石膏湿法烟气脱硫系统(wet flue gas desulfurization,WFGD)工作过程中浆液pH难以精准测量、不利于WFGD作业的问题,建立一种基于双向门控循环单元的脱硫系统pH预测模型。首先,对原始数据进行清洗和归一化处理;其次,基于最大信息系数分析得出13个特征值为输入变量,pH为输出变量,并建立浆液pH模型;最后,运行模型,并对结果进行评价。研究结果显示,与长短期记忆和门控循环相比,所选用的数学模型的平均绝对误差分别下降了11.95%、24.92%,均方根误差分别下降了10.64%、19.49%,决定系数分别提高了1.79%、3.08%。表明基于双向门控循环单元的pH预测模型具有较高的精确度和稳定性,具有工程应用价值,为现有脱硫塔pH预测模型提供了工程参考。

石灰石-石膏湿法烟气脱硫系统  /  双向门控循环单元神经网络  /  预测模型  /  浆液pH

Aiming to address the issue of accurately measuring the slurry pH value during the operation of limestone-gypsum wet flue gas desulfurization (WFGD) system, which hinders the efficient operation of WFGD, a prediction model for FGD system pH based on bi-directional gated recirculation unit(BiGRU) has been developed. Firstly, the raw data were cleaned and normalized. Secondly, based on the maximum information coefficient analysis, 13 characteristic values were obtained as input variables and pH as output variables, and a slurry pH model was established. Finally, the model was run and the results were evaluated. Compared to LSTM and GRU, the results indicate that the mean absolute error of this mathematical model decreases by 11.95% and 24.92%, while the root mean square error decreases by 10.64% and 19.49%. Additionally, the coefficient of determination improves by 1.79% and 3.08% respectively. This demonstrates that the BiGRU-based pH predictive model exhibits high accuracy and stability, making it valuable for engineering applications and providing an important reference for predicting pH models in existing desulfurization tower systems.

limestone-gypsum wet flue gas desulfurization system  /  BiGRU  /  prediction model  /  slurry pH
高钾, 田雪峰, 彭献永, 蒋甲丁, 徐敏. 基于双向门控循环单元的脱硫系统pH预测模型. 科学技术与工程, 2025 , 25 (13) : 5535 -5543 . DOI: 10.12404/j.issn.1671-1815.2404890
Jia GAO, Xue-feng TIAN, Xian-yong PENG, Jia-ding JIANG, Min XU. pH Prediction Model of Desulfurization System Based on Bi-directional Gated Recurrent Unit[J]. Science Technology and Engineering, 2025 , 25 (13) : 5535 -5543 . DOI: 10.12404/j.issn.1671-1815.2404890
石灰石-石膏湿法烟气脱硫技术是中国燃煤电厂主要的脱硫技术[1-3]。在此技术中,吸收塔中的浆液pH不仅影响脱硫效率,也决定吸收剂的用量[4]。此外,随着“双碳”目标的提出,中国大力推动新能源发展,建设新型电力系统,其中,煤电的角色从原来的基础电源向调节型、安全保障型电源转变。燃煤机组负荷变化频繁,脱硫系统SO2释放量波动加剧,加大吸收塔浆液pH控制难度[5-6]。因此,精确测量浆液pH极为关键。在实际操作环境下,实时监测pH的准确性常受到挑战,主要源于pH传感器在恶劣工作环境下的磨损,易造成测量不准确,从而影响了浆液pH,导致维持系统长期稳定运行比较困难[4-6]。为确保pH测量的精度足够高,理论上需在无干扰、稳定条件下进行,但现实难以达成,因此,发展pH新测量技术很有必要。
鉴于脱硫系统机理复杂且具有非线性和时变性,建立可靠的pH机理模型是一个挑战[6]。同时,随着人工智能和大数据技术的快速发展,数据驱动的预测技术已在各行各业展现出广泛的应用价值,脱硫系统也不例外。李建强等[7]提出了一种基于改进即时学习算法的脱硫系统吸收塔浆液pH软测量方法,仿真表明该模型通过考虑输入变量与输出变量之间的相关性,使学习集的选择更加合理,具有更好的预测效果。马双忱等[8]设计了一个深度神经网络模型,包括一个长短期记忆层、两个线性整流函数层和两个全连接层,对pH的预测值的均方根误差为0.094 7,优于深度BP的0.116。金秀章等[9]和柴晋等[1]从多方面分析了湿法脱硫系统自身缺陷和多种神经网络在脱硫系统预测方面的现状及价值,强调了长短时记忆网络(long short-term memory,LSTM)神经网络结构的优点,并仿真模拟,验证了LSTM的优越性。但随着技术不断进步,LSTM因参数众多导致训练效率受限,而门控循环单元(gate recurrent unit,GRU)神经网络在减少“门”数量的基础上,不仅提高了运算效率,还基本维持了与LSTM相似的性能特点[10]。郝晴等[11]使用GRU网络模型对吸收塔内的浆液pH进行预测,结果表明相比于反向传播神经网络模型、径向基函数神经网络、循环神经网络和长短期记忆神经网络,该模型精确度更高,实用性更强。尽管GRU模型的能力在捕捉历史信息方面表现出色,但它无法处理时间序列的双向性,导致预测结果还是不太精确[12]
因此,现提出选择能够有效处理时间序列数据前后依赖性的双向门控循环单元模型(bidirectional gate recurrent unit,BiGRU),建立脱硫系统浆液pH的软测量模型。首先对采集到的数据进行清洗,并用归一化和最大信息系数完成降维工作,然后建立基于BiGRU的脱硫系统pH预测模型,用大量数据进行训练,最后运行程序输出pH预测值,与LSTM、GRU相比,验证BiGRU模型的有效性。
石灰石-石膏湿法脱硫(wet flue gas desulfurization,WFGD)工艺采用石灰石作为主要脱硫材料,首先将其研磨成粉状,随之按既定比例与水混合形成浆液,或采用湿磨工艺直接融合二者。随后,制备好的浆液借助输送装置导入吸收塔的喷淋区域,经喷嘴雾化喷洒开来,覆盖塔内[13-14],如图1所示。这一过程涉及4个主要的化学反应阶段:吸收、溶解、氧化和结晶。其中,吸收和溶解过程受到浆液pH的影响较大,主要反应方程式如下。
SO2(g)+H2O→H2SO3
H2SO3→H++HS O 3 -
H++CaCO3→Ca2++HC O 3 -
Ca2++HS O 3 -+2H2O→CaSO3·2H2O+H+
H++HC O 3 -→H2CO3
H2CO3→CO2+H2O
吸收过程化学反应式为式(1)和式(2),溶解过程为式(3)~式(6)。观察到,氢离子(H+)在整个过程中扮演了至关重要的角色,可作为一种连接的媒介,具体可用pH反映。当pH升高,H+的浓度会降低,有利于吸收烟气中的二氧化硫,烟气脱硫率提高,但pH过高会降低石灰石的溶解度,阻碍钙离子的析出;反之,降低浆液pH,有助于钙离子的析出,但会抑制吸收塔内SO2的吸收效率,烟气脱硫率降低,且对脱硫设备的腐蚀增大[14]。此外,精确调控pH对于维持适宜的钙硫比率、提升石灰石使用效能及确保副产品石膏的品质具有重要作用。pH不当,易引发浆液沉积堵塞等问题,加速设备老化,加重电站运维负担及原料成本,同时,还可能促使脱硫后烟气SO2排放超标,超过环保标准[12-13]
因此,在燃煤机组快速调峰和系统延时性导致测量准确性下降的背景下,准确预测浆液pH尤为重要,为确保pH控制在最佳区间奠定基础。
与LSTM一样,GRU也是循环神经网络(recurrent neural network,RNN)的变体。为了解决LSTM在训练参数多和收敛速度慢的问题,Cho等[15]在2014年提出了门控循环单元神经网络(GRU)。其核心机制围绕两个关键门控运作:更新门与重置门。如图2所示,更新门决定过往时刻隐藏状态在当前状态中的保留比例,影响信息的延续性;重置门决定模型如何基于当前输入与前一状态进行信息的整合,从而适应序列中动态变化的需求[15-18]。这种设计使得GRU能够有效地调节信息的流动,既能保留长期依赖也能避免潜在的梯度消失问题。GRU前向传递过程具体如下。
步骤1 重置门计算。利用当前输入和前一隐藏状态来计算重置门的值。重置门决定从前一状态遗忘多少信息,越接近1,则表示过去信息越多。
$\boldsymbol{r}_{t}=\sigma\left(\boldsymbol{W}_{\mathrm{r}} \cdot\left[\boldsymbol{X}_{t}, \boldsymbol{h}_{t-1}\right]+\boldsymbol{b}_{\mathrm{r}}\right) $
$\sigma(x)=\frac{1}{1+\mathrm{e}^{-x}}$
式中:rt为重置门;Xt为当前输入;ht-1为上一时刻隐含信息;σ为sigmoid函数;[ ]表示矩阵堆叠;Wr为重置门的权重;br为重置门的偏置。
步骤2 更新门计算。同样利用当前输入和前一隐藏状态来确定更新门的值。这个门控制了多少之前的信息会被保留到当前时刻。zt越接近1,则表示现时刻更新信息越多。
$z_{t}=\sigma\left(\boldsymbol{W}_{z} \cdot\left[\boldsymbol{X}_{t}, \boldsymbol{h}_{t-1}\right]+\boldsymbol{b}_{z}\right)$
式(9)中:zt为更新门;Wz为更新门的权重;bz为更新门的偏置。
步骤3 候选隐藏状态。结合重置门的输出,计算一个候选隐藏状态。这个状态是基于当前的输入和经过重置门调整后的前一隐藏状态的组合。tanh函数可将数值变换到[-1,1]的区间内,得到候选隐藏状态内容hxt
$\boldsymbol{h}_{x t}=\tanh \left(\boldsymbol{W}_{\mathrm{h}} \cdot\left[\boldsymbol{X}_{t}, \boldsymbol{r}_{t} \otimes \boldsymbol{h}_{t-1}\right]+\boldsymbol{b}_{\mathrm{h}}\right)$
tanh(x)= e x - e - x e x + e - x
式中:tanh为双曲正切函数;Wh为候选隐藏状态的权重;bh为候选隐藏状态的偏置;⊗表示逐元素乘法。
步骤4 最终隐藏状态。使用更新门来混合前一隐藏状态和候选隐藏状态,生成当前的隐藏状态。将候选隐藏状态hxt与更新门zt作乘法,得到对现有信息进行选择性记忆;1-zt与上一时刻隐藏状态ht-1的乘积为对原本的记忆进行遗忘。两者之和为该时刻的隐藏状态ht,输入下一时刻,进行向前传递过程。这一步确保了网络可以根据需要保留旧的信息或添加新的信息。
$\boldsymbol{h}_{t}=\left(1-\boldsymbol{z}_{t}\right) \otimes \boldsymbol{h}_{t-1}+\boldsymbol{z}_{t} \otimes \boldsymbol{h}_{x t}$
为了使模型能够综合捕获序列中过去和未来信息,Chung等[19]设计了双向门控循环单元(bi-directional gated recurrent unit, BiGRU),如图 3 所示。该设计整合两个 GRU 单元,一个沿时间序列正向传播处理信息,另一个反向回溯,从而让模型同时拥有回顾过去与展望未来的能力,显著增强了时间序列数据分析的性能与预测精度。每个BiGRU单元在时刻t的输出,受到该时刻的偏置量、前向GRU输出和后向GRU输出的共同作用,公式[20]如下。
$\overrightarrow{\boldsymbol{h}}_{t}=\operatorname{GRU}\left(\boldsymbol{x}_{t}, \overrightarrow{\boldsymbol{h}}_{t-1}\right)$
$\overleftarrow{\boldsymbol{h}}_{t}=\operatorname{GRU}\left(\boldsymbol{x}_{t}, \overleftarrow{\boldsymbol{h}}_{t+1}\right)$
ht= m h t+ n h t+ct
式中:GRU(·)为GRU的计算过程; h tt时刻GRU前向隐藏层的输出值; h tt时刻GRU后向隐藏层的输出值;m为GRU前向隐藏层输出权重;n为GRU后向隐藏层输出权重;ctht对应的隐藏层偏置量。
通过浙江省某电厂2号机组分散控制系统,采集从2024年4月27日00:00到4月29日04:00的详细热力参数数据。所采集的数据集包括锅炉负荷、总风量、进口烟气流量及温度、浆液pH、浆液液位、进口SO2浓度在内的41项关键特征参数,每10 s取一次,共计10 080个数据点。初步分析,浆液循环泵、石膏排出泵、氧化风机电流等13个特征参数在观测期间未见变动,相关性较低,所以不予考虑,剩余28个特征值依然能有效反映系统运行情况。最后,将数据集划分为训练集、验证集(预测集),两者比例为8∶2。机组负荷和浆液pH原始数据分别如图4图5所示,可以看出,样本数据包含所有运行值。
分析28个特征参数发现数据集内不同变量之间的量级差异较大,若直接用原始数据训练模型可能导致模型稳定性差[21]。所以,首先对输入数据进行清洗,并通过式(16)进行线性归一化处理,使数据尺寸标准化[22]
Xn= X - X m i n X m a x - X m i n
式(16)中:Xn为归一化之后的数据;X为当前待归一化的数据;Xmin为待归一化总数据的最小值;Xmax为待归一化总数据的最大值。
可用于机器学习的特征选择方法很多,如:person、Spearman、阈值相关和最大信息系数等,经过简单比较多种方法的优缺点[23],选择最大信息系数,因为其具有普适性、公平性和对称性。
最大信息系数(maximal information coefficient, MIC) 是由David等[24]在 2011 年基于互信息理论提出的。该系数能在大型数据集中识别变量之间的线性或非线性关系,并作为评估两个变量相关性的标准。计算MIC的过程首先为两个变量的样本点绘制散点图,然后对散点图进行网格划分。通过计算各网格的边际概率密度和联合概率密度,进而得出互信息的值[24-25]。MIC首先涉及互信息,公式为
I(x;y)=∫p(x,y)log2 p ( x , y ) p ( x ) p ( y )dxdy
式(17)中:p(x,y)为xy之间的联合概率;p(x)和p(y)分别为xy的边缘概率密度函数。
联合概率密度计算通过网格划分实现,设一个已知数据集D,表达式为
D={(x1,y1),(x2,y2),…,(xm,ym)}
首先,将X轴和Y轴分别划分成ab个格子,形成网格G(a, b)。然后,绘制两个变量的散点图,并观察这些散点在网格G中的分布情况。在所有潜在的网格配置里,最高的互信息量就是最大信息系数的值。具体计算公式为
I'(D,x,y)=maxI(D|G)
如果有限的有序集合D的样本量为n,那么集合D中的变量XY的最大信息系数公式为
MIC(x;y)= m a x a × b < B ( n ) I ' ( D , x , y ) l o g 2 [ m i n ( a , b ) ]
式(20)中:ab分别为在xy方向上划分格子的个数;B(n)为网格划分的上限值。
在计算MIC时,B(n)的设定至关重要,它决定网格划分的搜索上界。若设定值过大,会把随机关系变量误显示为有相关性;若设定值过小,可能只能捕捉到两个变量间的简单关系,忽略更复杂的相互作用。为确保MIC的有效性和可靠性,常用设定方法是将B(n)设为n的0.6次幂(即n0.6),平衡了详细程度和计算的可行性[26]
尽管对原始41维特征值初步处理到28维,但维数仍然很大,既不便于计算,又可能会产生“维数灾难”的问题,选择MIC继续降维,结果如图6所示,其中有13个特征值与pH的相关系数大于60%,如表1所示。
图6表1分析出,输入变量排序能够解释脱硫系统pH的变化机理。再循环泵5电流值与浆液pH影响最大,因为循环泵控制浆液的流量,且浆液程碱性,流量增大,pH升高;流量减小,pH降低。pH影响到实际脱硫效率,反之,实际脱硫效率会反映pH的高低,从它们之间的相关性为0.866就能看出。吸收塔给浆密度高低,直接反映pH的高低,因为石灰石溶于水呈碱性,而石灰石耗量和吸收塔给浆体积流量决定石灰石给浆密度,共同影响pH。FGD出口粉尘含量与浆液pH之间存在的较大关系,因为烟气中的粉尘会附着在浆液中的石灰石颗粒表面,使石灰石的溶解受限,进而影响浆液pH。高温烟气使SO2与脱硫剂接触时的传质效果变差,导致SO2吸收不充分,消耗更多碱性物质以保持脱硫效率,从而降低浆液pH;高温阻碍石膏正常结晶,促进亚硫酸盐氧化产生硫酸,也会导致浆液酸度增加。所以,烟气温度会通过多种途径影响脱硫系统中SO2的吸收效率和化学反应平衡,间接导致浆液pH发生变化。机组负荷增加通常伴随SO2排放增多,间接影响浆液pH。锅炉总风量增大可能缩短烟气在脱硫塔内的停留时间,降低SO2吸收效率,可能导致浆液pH升高;风量减小时,可能导致pH降低。实践生产中,负荷与风量变化时,需相应调整浆液循环量、喷淋密度等,确保烟气与浆液充分接触。若调整不当,可能加速碱性物质消耗,引发pH下降。若其他条件不变,当FGD进口烟气含氧量不断增加时,吸收塔内进行的化学反应速率得以加快,降低浆液pH。FGD进口烟气SO2浓度变化将会直接影响吸收塔内浆液pH的变化,烟气SO2中浓度越高,则浆液pH越低。
通过MIC对28个特征值降维,并对影响浆液pH因素进行机理分析,选择前13个变量作为模型输入变量,浆液pH作为输出量。
采用BiGRU模型进行pH预测,模型结构共分为输入层、隐藏层、全连接层和输出层,如图7所示。具体步骤如下。
步骤1 对原始数据进行清洗、归一化处理,并通过MIC降维,然后划分数据为训练集和测试集。
步骤2 设计并建立BiGRU网络模型,确定网络结构,设置训练迭代次数、学习率等参数,并选择合适的优化算法和损失函数。
步骤3 采用选定的优化算法对网络模型参数进行优化,建立脱硫系统浆液pH预测模型,并通过反复训练直至达到预期效果。
步骤4 将测试集样本数据输入优化后的模型,获取预测结果并进行反归一化处理,然后利用评价函数对预测结果进行分析和评价。
BiGRU模型使用MATLAB R2023b编程实现,所使用的电脑搭载AMD Ryzen5 3500U处理器,主频为1.6 GHz,配备12.0 GB内存,显卡的型号Radeon Vega,16 GB。输入层节点个数为13,输出层节点个数为1,隐藏层神经元个数为128,隐藏层层数设置为2,学习步长为1,小批量样本数量MiniBatchSize为256,初始学习率设为0.01,批处理量设为200,且采用自适应学习率的Adam优化算法。调试过程如表2~表4所示,以平均绝对误差(mean absolute error,MAE)、均方根误差(root mean squared error,RMSE)为评价指标。
表2中,当MiniBatchSize为256时,训练集上的平均绝对误差和均方根误差最小,测试集上的误差也小于其他3个参数值对应的误差。所以选定MiniBatchSize为256。
表3中,对比不同最大训练代数MaxEpochs所对应的误差,当MaxEpochs为200时,训练集和测试集效果都较好,故选取MaxEpochs参数值为200。
表4所示,对比可以看出,当初始学习率InitialLearnRate为0.01时,训练集和测试集的误差相比于其他参数时较小,则选取0.01为InitialLearnRate参数值。
BiGRU网络模型对浆液pH预测的评价指标采用平均绝对误差(MAE)、均方根误差(RMSE)和决定系数R2,如式(21)~式(23)所示[25]。MAE是模型预测值与模型真实值之间的平均绝对误差,RMSE反映泛化误差水平,决定系数R2对计算结果进行了归一化处理,弥补了MAE和RMSE在评价不同模型、不同量纲数据时的不足。MAE、RMSE越接近0,R2越接近1,代表模型预测越准确。
MAE= 1 n i = 1 n y i - y i
RMSE= 1 n i = 1 n ( y i - y i ) 2
R2=1- ( y i - y i ) 2 ( y i - y - i ) 2
式中:yi为pH的真实值; y i为pH的预测值; y - i为pH的平均值;n为预测集数据个数。
为验证MIC算法对变量筛选的有效性,现建立对比模型实验:分别以全部未筛选的28个特征变量和经过MIC筛选的13个重点特征变量为模型输入,采用以上超参数并用同一个模型分别运行程序,对浆液pH进行预测,预测结果如图8所示,误差如图9所示,评价指标值和运行时间如表5所示。
图8图9表5可知,特征变量经过筛选后,模型预测精度获得了提升,运行速度更快。与筛选前模型预测结果相比,筛选后模型的MAE下降了13.98%,RMSE下降了13.81%,R2提高了1.70%,运行时间下降了11.92%。特征变量经过筛选后,预测值波动幅度变小,稳定性增强:筛选后误差稳定在区间(-0.05, 0.05),而筛选前误差频繁突破±0.05,个别情况还突破±0.1;当浆液pH变化时,BiGRU预测值与实际值能实现紧密跟随,但是随着时间增长,筛选前曲线更容易偏离真实值曲线,出现骤升或骤降。实验表明,原始数据特征变量太多,互相影响,相互作用,存在冗余关系,这会影响模型预测精度,所以有效的筛选特征变量一定程度可以上减少模型复杂度,提高模型预测精度,降低模型运行时间。
为验证模型的有效性和准确性,在使用同一个数据集的基础上,分别使用BiGRU、LSTM和GRU神经网络模型对WFGD系统的浆液pH进行预测,结果和真实值如图10所示,模型的预测误差如图11所示。为体现所列样本具有代表性,每隔300序列选取一个原始数据,共选择7个,制作各模型的部分预测数据,其和评价指标值如表6所示。
图10图11可以看出,在WFGD系统的浆液pH预测模型中,当浆液pH保持不变时,GRU和LSTM模型预测值频繁、剧烈波动,且随着时间的拉长,有下降趋势,且GRU比LSTM预测值波动幅度大,而BiGRU预测值波动不大,稳定性高,误差稳定在区间(-0.05,0.05);当浆液pH上升或者下降时,BiGRU预测值与实际值紧密跟随,且增加后迅速保持稳定,而GRU和LSTM模型预测值有时滞后实际值,且波动幅度大,尤其是GRU模型,误差经常突破±0.05,甚至±0.1。对比三种模型预测结果,BiGRU的预测值具有更高的稳定性和准确度,趋势跟随紧密,误差小,验证了其在数据挖掘和双向时间序列处理上的优势。
表6可以得出,对于任何一种模型,7个样本点预测值与真实数据的差值都大小不同,BiGRU的误差最小,只有第400个样本点的误差超过±0.02,为-0.021 611;LSTM预测的7个样本里,超过±0.02有4个,超过±0.03有2个,最大误差为0.037 262 9;关于GRU预测结果,超过±0.02有5个,超过±0.03有4个,超过±0.04有2个,最大误差为-0.046 210 4,这表明BiGRU预测结果最接近真实值。BiGRU模型的平均绝对误差为0.013 904、均方根误差为0.018 311,与LSTM相比,分别下降了11.95%、10.64%;与GRU相比,分别下降了24.92%、19.49%。BiGRU模型的R2为0.988 71,与LSTM和GRU相比,分别提高1.79%、3.08%。这说明BiGRU模型考虑双向信息对实时数据的影响,比起LSTM和GRU模型进一步提高了模型训练的精度和稳定性,更具有工程应用价值。
(1)石灰石-石膏湿法脱硫系统的机理复杂,影响浆液pH因素众多,且相互影响,存在非常复杂的非线性关系,采用MIC方法选取13个特征值,取得了较好的效果。
(2)基于BiGRU模型的石灰石-石膏湿法脱硫系统浆液pH预测模型能够将该机组的运行数据同时间序列相关联,并快速分析和训练,准确地预测浆液pH,为下一步pH控制奠定基础。
(3)与LSTM和GRU模型相比,BiGRU模型的平均绝对误差和均方根误差均有所降低,而R2有所提高,预测结果较好,泛化能力较强,这说明引入了逆向信息的反馈能提高预测精度和稳定性。
  • 国家自然科学基金(72361033)
  • 中国华能集团公司总部科技资助项目(HNKJ23-HF31)
参考文献 引证文献
排序方式:
[1]
柴晋, 乔加飞, 孙灏, 等. 神经网络算法在脱硫系统优化中的应用进展[J]. 洁净煤技术, 2021, 27(2): 27-32.
Chai Jin, Qiao Jiafei, Sun Hao, et al. Progress in the application of neural network algorithm in the optimization of desulfurization system[J]. Clean Coal Technology, 2021, 27(2): 27-32.
[2]
Tang L, Qu J B, Mi Z F, et al. Substantial emission reductions from Chinese power plants after the introduction of ultra-low emissions standards[J]. Nature Energy, 2019, 4(11): 929-938.
[3]
孔若琪, 崔琳, 董勇. 机器学习算法在脱硫系统智能运行及优化中的应用[J]. 洁净煤技术, 2023, 29(S2): 406-414.
Kong Ruoqi, Cui Lin, Dong Yong. Application of machine learning algorithm in intelligent operation and optimization of desulfurization system[J]. Clean Coal Technology, 2023, 29(S2): 406-414.
[4]
康英伟, 孙智滨, 常俊. 基于改进差分进化算法的湿法烟气脱硫系统模型辨识[J]. 动力工程学报, 2019, 39(10): 834-839.
Kang Yingwei, Sun Zhibin, Chang Jun. Model identification of wet flue gas desulfurization system based on improved differential evolutionary algorithm[J]. Journal of Power Engineering, 2019, 39(10): 834-839.
[5]
陈浩, 骆仲泱, 江建平, 等. 电厂湿法烟气脱硫颗粒物排放特性的实验研究[J]. 动力工程学报, 2017, 37(11): 925-930, 937.
Chen Hao, Luo Zhongyang, Jiang Jianping, et al. Experimental study on the emission characteristics of particulate matter from wet flue gas desulfurization in a power plant[J]. Journal of Power Engineering, 2017, 37(11): 925-930, 937.
[6]
Liu G F, Zhang Y, Shen D K, et al. Anticipatory NH3 injection control for SCR system based on the prediction of the inlet NOx concentration[J]. Journal of the Energy Institute, 2021, 94: 167-175.
[7]
李建强, 杨红, 牛成林. 基于改进即时学习算法的吸收塔pH软测量技术研究[J]. 电力科学与工程, 2021, 37(8): 60-66.
Li Jianqiang, Yang Hong, Niu Chenglin. Research on soft measurement technology of pH value in absorption tower based on improved instant learning algorithm[J]. Electric Power Science and Engineering, 2021, 37(8): 60-66.
[8]
马双忱, 林宸雨, 周权. 基于深度神经网络的脱硫系统预测模型及应用[J]. 化工进展, 2021, 40(3): 1689-1698.
Ma Shuangchen, Lin Chenyu, Zhou Quan. Prediction model and application of desulfurization system based on deep neural network[J]. Advances in Chemical Engineering, 2021, 40(3): 1689-1698.
[9]
金秀章, 景昊. 基于 LSTM 的湿法烟气脱硫浆液pH建模[J]. 信息技术与网络安全, 2020, 39(8): 62-66.
Jin Xiuzhang, Jing Hao. LSTM-based pH modeling of wet flue gas desulfurization slurry[J]. Information Technology and Network Security, 2020, 39(8): 62-66.
[10]
Gupta U, Bhattacharjee V, Bishnu S. Stock Net-GRU based stock index prediction[J]. Expert Systems with Applications, 2022, 207: 117986.
[11]
郝晴, 迟涛, 于政军, 等. 基于门控循环单元神经网络的浆液pH的预测建模[J]. 科学技术与工程, 2023, 23(18): 7824-7830.
Hao Qing, Chi Tao, Yu Zhengjun, et al. Predictive modeling of slurry pH based on gated recurrent unit neural network[J]. Science Technology and Engineering, 2023, 23(18): 7824-7830.
[12]
Li W D, Gao H H, Cai W F. Scaling analysis and preventive measures of limestone-gypsum wet flue gas desulfurization[J]. Power Generation Technology, 2019, 40(1): 51-55.
[13]
曹建宗, 刘琦, 陈文通, 等. 典型湿法脱硫系统存在的问题及人工智能在优化运行中的应用[J]. 化工进展, 2020, 39(S1): 242-249.
Cao Jianzong, Liu Qi, Chen Wentong, et al. Problems of typical wet desulfurization systems and the application of artificial intelligence in optimized operation[J]. Progress in Chemical Engineering, 2020, 39(S1): 242-249.
[14]
董家伟. 湿法烟气脱硫pH的模型预测控制[D]. 北京: 北京工业大学, 2022.
Dong Jiawei. Model predictive control of pH in wet flue gas desulfurization[D]. Beijing: Beijing Institute of Technology, 2022.
[15]
Cho K, Van Merriënboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder forstatistical machine translation[J]. ArXiv Preprint ArXiv, 2014: 1406. 1078.
[16]
Wang D, Mao K. Multimodal object classification using bidirectional gated recurrent unit networks[C]// 2018 IEEE Third International Conference on Data Science in Cyberspace (DSC). Guangzhou: IEEE, 2018: 685-690.
[17]
赵兵, 王增平, 纪维佳, 等. 基于注意力机制的CNN-GRU短期电力负荷预测方法[J]. 电网技术, 2019, 43(12): 4370-4376.
Zhao Bing, Wang Zengping, Ji Weijia, et al. CNN-GRU short-term power load forecasting method based on attention mechanism[J]. Grid Technology, 2019, 43(12): 4370-4376.
[18]
蒋星明, 曹顺安, 王民军, 等. 双向门控循环神经网络的SO2排放浓度预测模型[J]. 应用化工, 2021, 50(12): 3519-3523.
Jiang Xingming, Cao Shun'an, Wang Minjun, et al. SO2 emission concentration prediction modeling by two-way gated recurrent neural network[J]. Applied Chemical Engineering, 2021, 50(12): 3519-3523.
[19]
Chung J, Gulcehre C, Cho K H, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J]. ArXiv Preprint ArXiv, 2014: 1412. 3555.
[20]
许崇轩. 基于相空间重构和双向门控循环单元神经网络的海水指标预测研究[D]. 秦皇岛: 燕山大学, 2023.
Xu Chongxuan. Research on seawater index prediction based on phase space reconstruction and two-way gated recurrent unit neural network[D]. Qinhuangdao: Yanshan University, 2023.
[21]
杨寒雨, 赵晓永, 王磊. 数据归一化方法综述[J]. 计算机工程与应用, 2023, 59(3): 13-22.
Yang Hanyu, Zhao Xiaoyong, Wang Lei. A review of data normalization methods[J]. Computer Engineering and Applications, 2023, 59(3): 13-22.
[22]
孙坤, 尹晓红. 基于数据去噪和CNN-BiGRU的SO2排放预测[J]. 电子测量技术, 2023, 46(13): 66-72.
Sun Kun, Yin Xiaohong. SO2 emission prediction based on data denoising and CNN-BiGRU[J]. Electronic Measurement Technology, 2023, 46(13): 66-72.
[23]
顾俊发, 许明阳, 马方圆, 等. 基于MIC的支持向量回归及其在化工过程中的应用[J]. 化工学报, 2021, 72(3): 1480-1486.
Gu Junfa, Xu Mingyang, Ma Fangyuan, et al. MIC-based support vector regression and its application to chemical processes[J]. Journal of Chemical Engineering, 2021, 72(3): 1480-1486.
[24]
David N, Reshef Y A, Finucane H K, et al. Detecting novel associntionsin large data sets[J]. Science, 2011, 334(6062): 1518-1524.
[25]
吴磊, 康英伟. 基于长短时记忆-反向传播级联模型的湿法脱硫系统性能预测[J]. 科学技术与工程, 2021, 21(29): 12616-12622.
Wu Lei, Kang Yingwei. Performance prediction of wet desulfurization system based on long-short-term memory-back propagation cascade model[J]. Science Technology and Engineering, 2021, 21(29): 12616-12622.
2025年第25卷第13期
PDF下载
335
134
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2404890
  • 接收时间:2024-06-30
  • 首发时间:2025-07-09
  • 出版时间:2025-05-08
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-06-30
  • 修回日期:2025-02-07
基金
国家自然科学基金(72361033)
中国华能集团公司总部科技资助项目(HNKJ23-HF31)
作者信息
    1 中国矿业大学低碳能源与动力工程学院, 徐州 221116
    2 新疆工程学院能源工程学院, 乌鲁木齐 830023
    3 中煤玉电(玉环)能源开发有限公司, 台州 317604

通讯作者:

* 彭献永(1975—),男,汉族,江苏徐州人,博士,副教授,硕士研究生导师。研究方向:智能发电与综合智慧能源。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2404890
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