面向炼化生产运行的高质量多源时序数据集构建方法研究

刘晋瑞; 刘雪鹏; 刘速; 吕庭龙; 戴赟; 王振雄

doi:10.11949/0438-1157.20260411

您当前的位置：

首页 >

文章列表页 >

面向炼化生产运行的高质量多源时序数据集构建方法研究

其他 | 更新时间：2026-05-06

- 面向炼化生产运行的高质量多源时序数据集构建方法研究
- Research on construction methods for high-quality multi-source time series datasets for refining and chemical production operations
- 化工学报 2026年
- 作者机构：
  
  1.中石油（北京）数智研究院有限公司，北京 102206
  2.中国石油天然气股份有限公司长庆石化分公司，陕西咸阳 712042
  3.中国石油天然气股份有限公司兰州石化分公司，甘肃兰州 730060
- 作者简介：
  
  刘晋瑞（1999—），女，硕士，工程师，liu_jinrui@petrochina.com.cn
  刘雪鹏（1983—），男，博士，高级工程师，liuxuepeng01@petrochina.com.cn
- 基金信息：
  
  中国石油天然气股份有限公司科技项目(2025ZG86)
- DOI：10.11949/0438-1157.20260411
  中图分类号： TQ 056.1;TP 391
- 收稿：2026-03-27，
  
  修回：2026-05-06，
  
  录用：2026-05-06，
- 稿件说明：
移动端阅览
刘晋瑞, 刘雪鹏, 刘速, 等. 面向炼化生产运行的高质量多源时序数据集构建方法研究[J/OL]. 化工学报, 2026.

LIU Jinrui, LIU Xuepeng, LIU Su, et al. Research on construction methods for high-quality multi-source time series datasets for refining and chemical production operations[J/OL]. CIESC Journal, 2026.
刘晋瑞, 刘雪鹏, 刘速, 等. 面向炼化生产运行的高质量多源时序数据集构建方法研究[J/OL]. 化工学报, 2026. DOI： 10.11949/0438-1157.20260411.

LIU Jinrui, LIU Xuepeng, LIU Su, et al. Research on construction methods for high-quality multi-source time series datasets for refining and chemical production operations[J/OL]. CIESC Journal, 2026. DOI： 10.11949/0438-1157.20260411.

摘要

炼化生产过程中的 DCS、LIMS 与 MES 数据是支撑时序大模型训练与智能化应用的关键前提。针对此类数据普遍存在的结构不一致、质量参差、时间错配和语义关系割裂等问题，本文提出一套面向炼化生产运行的高质量多源时序数据集构建方法。该方法涵盖数据采集、存储、标准化、质量评估与处理分析五个核心环节：首先，基于分层分类目录结构实现多源数据的高效组织；其次，定义DCS、LIMS和MES数据统一时间戳与变量标识的标准化格式；进而，构建包含完整性、规范性、合理性与稳定性四个维度的可量化质量评估体系；最后，采用时间对齐与语义映射策略，有效解决不同来源数据在时间粒度与语义表达上的差异问题。本研究通过所提方法，构建了一套高质量炼化生产运行时序数据集，实现数据集的结构统一、质量可控和语义协同，为炼化行业数据资产化与智能化升级提供了可复用的数据构建范式。

Abstract

The DCS(Distributed Control System)

LIMS(Laboratory Information Management System)

and MES(Manufacturing Execution System) data in the refining and chemical production process are crucial prerequisites for supporting the training of large time-series models and intelligent applications. Addressing the prevalent issues in such data

such as inconsistent structures

varying quality

time mismatches

and fragmented semantic relationships

this paper proposes a method for constructing a high-quality multi-source time-series dataset tailored for refining and chemical production operations. This method encompasses five core aspects: data collection

storage

standardization

quality assessment

and processing analysis. Firstly

it achieves efficient organization of multi-source data based on a hierarchical classification directory structure. Secondly

it defines standardized formats for unified timestamps and variable identifiers for DCS

LIMS

and MES data. Furthermore

it constructs a quantifiable quality assessment system encompassing four dimensions: completeness

normalization

rationality

and stability. Lastly

it employs time alignment and semantic mapping strategies to effectively address the differences in time granularity and semantic expression among data from different sources. Through the proposed method

this study constructs a high-quality time-series dataset for refining and chemical production operations

achieving unified structure

controllable quality

and semantic coordination of the dataset. This provides a reusable data construction paradigm for the data assetization and intelligent upgrading of the refining and chemical industry.

关键词

Keywords

references

国家数据局 , 中共中央网络安全和信息化委员会办公室 , 工业和信息化部 , 等 . “ 数据要素×”三年行动计划（ 2024 — 2026 年）[Z ] . 2024 .

National Data Administration , Office of the Central Leading Group for Cyberspace Affairs , Ministry of Industry and Information Technology , etc . " Data Element ×" Three-Year Action Plan (2024-2026) [Z ] . 2024

Liang W X , Tadesse G A , Ho D , et al . Advances, challenges and opportunities in creating data for trustworthy AI [J ] . Nature Machine Intelligence , 2022 , 4 ( 8 ): 669 - 677 .

Liu Y , Cao J H , Liu C Y , et al . Datasets for large language models: a comprehensive survey [J ] . Artificial Intelligence Review , 2025 , 58 ( 12 ): 403

Gong Y D , Liu G Z , Xue Y Z , et al . A survey on dataset quality in machine learning [J ] . Information and Software Technology , 2023 , 162 : 107268 .

Pani A K , Soofastaei A . Designing intelligence: harnessing soft sensors and advanced analytics in petroleum refining for Industry 4.0 [M ] // Advanced Analytics for Industry 4.0 . Boca Raton : CRC Press , 2025 : 83 - 116 .

罗梦杰 , 赵云鹏 , 张梦轩 , 等 . 大模型时代的石油化工行业高质量数据集: 挑战与机遇 [J ] . 化工学报 , 2026 , 77 ( 1 ): 435 - 447, F 0004 .

Luo M J , Zhao Y P , Zhang M X , et al . High-quality datasets for petrochemical industry in era of large language models: challenges and opportunities [J ] . CIESC Journal , 2026 , 77 ( 1 ): 435 - 447, F 0004 .

Wang C J , Zhu W B , Gao B B , et al . Real-IAD: a real-world multi-view dataset for benchmarking versatile industrial anomaly detection [C ] // 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . June 16-22, 2024 , Seattle, WA, USA. IEEE , 2024 : 22883 - 22892 .

Onyeke F O , Odujobi O , Adikwu F E , et al . Advancements in the integration and optimization of control systems: Overcoming challenges in DCS, SIS, and PLC deployments for refinery automation [J ] . Open Access Research Journal of Multidisciplinary Studies , 2022 , 4 ( 2 ): 94 - 101 .

Andrade J M , Prada D , Muniategui S , et al . A laboratory information management system in a petrochemical company: an effective information solution [J ] . Chemometrics and Intelligent Laboratory Systems , 1993 , 21 ( 1 ): 35 - 49 .[LinkOut ]

Yuan Z H , Qin W Z , Zhao J S . Smart manufacturing for the oil refining and petrochemical industry [J ] . Engineering , 2017 , 3 ( 2 ): 179 - 182 .

郭佳佳 . 基于元数据的石油化工行业数据治理 [J ] . 石油知识 , 2024 ( 6 ): 46 - 47 .

Guo J J . Data governance of petrochemical industry based on metadata [J ] . Petroleum Knowledge , 2024 ( 6 ): 46 - 47 .

Wu Q Y , Yao D , Liu Y , et al . Research on construction methods of high-quality multimodal datasets in industrial manufacturing [C ] // 2025 IEEE International Conference on Blockchain (Blockchain) . October 30 - November 2 , 2025 , Zhengzhou, China . IEEE , 2025: 241 - 246 .

Duan R , Deng H , Tian M , et al . SODA: a large-scale open site object detection dataset for deep learning in construction [J ] . Automation in Construction , 2022 , 142 : 104499 .

Li J Y , Tang J Z , Zhao X Z , et al . A large-scale, high-quality dataset for lithology identification: Construction and applications [J ] . Petroleum Science , 2025 , 22 ( 8 ): 3207 - 3228 .

周军 , 石玉江 , 张娟 , 等 . 统一测井数据库建设与应用 [J ] . 测井技术 , 2022 , 46 ( 6 ): 757 - 761 .

Zhou J , Shi Y J , Zhang J , et al . Construction and application of unified logging database [J ] . Well Logging Technology , 2022 , 46 ( 6 ): 757 - 761 .

李梁 , 姜辉 , 宋一纯 , 等 . AI场景下数据治理体系建设和实施策略 [J ] . 管道保护 , 2025 , 2 ( 5 ): 78 - 82, 90 .

Li L , Jiang H , Song Y C , et al . Construction and implementation strategy of data governance system in AI scenario [J ] . Pipeline Protection , 2025 , 2 ( 5 ): 78 - 82, 90 .

刘珊珊 . 大语言模型在勘探开发数据治理中应用研究 [J ] . 北京石油化工学院学报 , 2025 , 33 ( 2 ): 64 - 72 .

Liu S S . Exploration of the application of large language models in exploration and development data governance [J ] . Journal of Beijing Institute of Petrochemical Technology , 2025 , 33 ( 2 ): 64 - 72 .

Shu L , Mukherjee M , Pecht M , et al . Challenges and research issues of data management in IoT for large-scale petrochemical plants [J ] . IEEE Systems Journal , 2018 , 12 ( 3 ): 2509 - 2523 .

Mohammed S , Budach L , Feuerpfeil M , et al . The effects of data quality on machine learning performance on tabular data [EB/OL ] . 2022 : arXiv: 2207.14529 .

Johnson D H . Signal-to-noise ratio [J ] . Scholarpedia , 2006 , 1 ( 12 ): 2088 .

Kini K R , Madakyaru M . Improved process monitoring scheme using multi-scale independent component analysis [J ] . Arabian Journal for Science and Engineering , 2022 , 47 ( 5 ): 5985 - 6000 .

Kashpruk N , Piskor-Ignatowicz C , Baranowski J . Time series prediction in industry 4.0: a comprehensive review and prospects for future advancements [J ] . Applied Sciences , 2023 , 13 ( 22 ): 12374 .

Bousetouane F . Agentic systems: a guide to transforming industries with vertical AI agents [EB/OL ] . 2025 : arXiv: 2501.00881 .

Liang G N , Tong Q Q . LLM-powered AI agent systems and their applications in industry [C ] // 2025 IEEE World AI IoT Congress (AIIoT) . May 28-30, 2025 , Seattle, WA, USA. IEEE , 2025 : 463 - 471 .

Ahmed S , Nielsen I E , Tripathi A , et al . Transformers in time-series analysis: A tutorial [J ] . Circuits, systems, and signal processing , 2023 , 42 ( 12 ): 7433 - 7466 .

Shi X M , Wang S Y , Nie Y Q , et al . Time-MoE: billion-scale time series foundation models with mixture of experts [EB/OL ] . 2024 : arXiv: 2409.16040 .

Gao S H , Hartvigsen T , Koker T , et al . UniTS: a unified multi-task time series model [C ] // Advances in Neural Information Processing Systems 37 . December 10 - 15 , 2024 . Vancouver, BC, Canada. Neural Information Processing Systems Foundation, Inc . (NeurIPS), 2024 : 140589 - 140631 .

Liu Y , Zhang H R , Li C Y , et al . Timer: transformers for time series analysis at scale [J ] . arXiv preprint , 2024 , arXiv: 2402.02368 .

Saghir H , Ahmad I , Kano M , et al . Prediction and optimisation of gasoline quality in petroleum refining: The use of machine learning model as a surrogate in optimisation framework [J ] . CAAI Transactions on Intelligence Technology , 2024 , 9 ( 5 ): 1185 - 1198 .

Gholamzadeh E , Shokri A , Ghaemi A , et al . Optimization of energy consumption in the gas refinery sweetening unit using RSM, ANN, and Aspen HYSYS [J ] . Scientific Reports , 2025 , 15 : 38763 .

Tian W D , Wang S C , Sun S L , et al . Intelligent prediction and early warning of abnormal conditions for fluid catalytic cracking process [J ] . Chemical Engineering Research and Design , 2022 , 181 : 304 - 320 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

大模型时代的石油化工行业高质量数据集：挑战与机遇

面向石油化工难分离体系的分离技术研究进展

功热集成中压力操作路径和换热匹配的同步优化

ORC发电系统工艺过程预警诊断技术研究与展望

石化行业产品生命周期碳足迹评价研究现状及展望