DC-4: Jingkang Liang

Contact

E-mail: jinlia@kt.dtu.dk

💼LinkedIn 🆔ORCID 🏛️ResearchGate 📖Google Scholar

Project

Statistical machine learning and big data analytics for online risk monitoring and fault diagnosis

Host organization

DTU

Supervisors

Prof. Gürkan Sin (Main, DTU); Prof. V. Venkatasubramanian (co-supervisor, COL); Dr. Claus Myllerup (co-supervisor, KAIROS)

Objectives

  • To develop a deep learning-based algorithm with uncertainty estimation for fault diagnosis in chemical processes
  • To explore the potential of cutting-edge artificial intelligence methodologies in enhancing safety in chemical processes
  • To develop high-quality dataset for training fault detection and diagnosis algorithms in chemical processes
  • To apply the developed methods to several relevant case studies, with a particular focus on the hydrogen industry

 

Project Description

My project (ProSafe DC-4), titled "Statistical Machine Learning and Big Data Analytics for Online Risk Monitoring and Fault Diagnosis," is to develop advanced data-driven methodologies for real-time fault diagnosis in complex industrial processes. As the industrial processes become increasingly complex and larger in scale, traditional methods for online risk monitoring and fault diagnosis often fall short when dealing with high-dimensional, heterogeneous data streams, resulting in delayed detection and response to system anomalies. In contrast, statistical machine learning and big data analytics, including deep learning-based methods, present promising solutions to these challenges by enabling more accurate and timely fault detection, thereby enhancing system reliability and safety.

Three key milestones define the trajectory of my project. First, it will focus on creating novel fault diagnosis models using deep learning and uncertainty estimation, enhancing the precision and reliability of fault diagnosis in complex process systems. Second, in light of the emerging capabilities of generative AI, the project will investigate and leverage advanced artificial intelligence techniques to elevate the safety of chemical processes, pushing the boundaries of current monitoring systems. Lastly, a critical component involves creating comprehensive, high-quality datasets specifically tailored for training fault detection and diagnosis algorithms. Additionally, the research will implement these advanced methodologies in several case studies, with a particular focus on a hydrogen liquefaction and processing facility.

In conclusion, my project aims to develop advanced statistical machine learning and big data analytics methodologies for real-time fault diagnosis in complex industrial processes, enhancing precision, reliability, and operational safety. By integrating these innovative methods into a hydrogen liquefaction and processing facility, the research seeks to validate their practical applicability and contribute significantly to the advancement of process systems engineering. 

Relevant Background

  • B.Sc.: Bachelor’s Degree in Mechanical Engineering, South China University of Technology, 2017-2021
  • M.Sc.: Master’s Degree in Mechanical Engineering – Intelligent Fault Diagnosis for rotating machinery using lightweight model and Automated Machine Learning, South China University of Technology, 2021-2024
  • Exchange study at Politecnico di Torino (PoliTO)

Publications

  1. Liang, Jingkang, et al. "Intelligent fault diagnosis of rotating machinery using lightweight network with modified tree‐structured parzen estimators." IET Collaborative Intelligent Manufacturing 4.3 (2022): 194-207. https://doi.org/10.1049/cim2.12055
  2. J. Liang, Z. Chen, J. Chen, J. Li, R. Huang and W. Li, "Resource-Efficient Network for Intelligent Fault Diagnosis Using Tree-Structured Parzen Estimator," 2023 7th International Conference on System Reliability and Safety (ICSRS), Bologna, Italy, 2023, pp. 585-589, doi: 10.1109/ICSRS59833.2023.10381129.
  3. Liang, J., Liao, Y., Li, W. (2023). Differentiable Architecture Searched Network with Tree-Structured Parzen Estimators for Rotating Machinery Fault Diagnosis. In: Zhang, H., Feng, G., Wang, H., Gu, F., Sinha, J.K. (eds) Proceedings of IncoME-VI and TEPEN 2021. Mechanisms and Machine Science, vol 117. Springer, Cham. https://doi.org/10.1007/978-3-030-99075-6_77