Education

M.S. in Hunan University, Civil Engineering (ARWU Top 15 Academic Subjects Worldwide), 2024

B.S. in Chongqing Jiaotong University, Civil Engineering, 2021

Publication

1. Ren, Y., Zhang, C.*, Zhu, M., Chen, R., Wang, J. (2023). Significance and formulation of ground loss in tunnelinginduced settlement prediction: a data-driven study. Acta Geotechnica, 1-16. (Impact Factor : 6.0)
2. He, S.*, Ren, Y., Liu, H., Liang, B., Du, G. (2022). A novel tunnel lighting method aided by highly diffuse reflective materials on the sidewall: Theory and practice. Tunnelling and Underground Space Technology, 122, 104336. (Impact Factor : 6.9)
3. Geng, Z., Zhang, C.*, Ren, Y., Zhu, M., Chen, R., Cheng, H. (2023). A Kriging-Random Forest Hybrid Model for Real-time Ground Property Prediction during Earth Pressure Balance Shield Tunneling. arXiv preprint arXiv:2305.05128. (Under revision)
4. Ren, Y., Zhang, C.*. (2023). A quality index for construction big data in EPB shield tunneling. (In preparing)

*:Supervisor

Experience

• Physics-informed inverse modeling for SWRC and HCF
  • Specify an inverse PINN for identifying SWRC and HCF: Encode the Richardson-Richards equation (RRE) with unknown parameters that determine the hydraulic properties into the loss function of a neural network, yielding a novel PINN for identifying the constitutive models
  • Validation: Simulated data and experiment data are utilized for demonstrating the effectiveness of the proposed framework

• Quality index for construction big data in EPB shield tunneling
  • Quality index development: Pioneer the first quality measure for construction big data in EPB shield tunneling, which consists of three components, i.e., representativeness, diversity, and informativeness
  • Verify the effectiveness of the proposed quality index: The quality index of data is quantitatively compared with the performance of the models developed by data via a correlation analysis, the high R-values that more than 0.9 confirms the effectiveness of the quality index

• Data-driven modeling for ground loss in EPB shield tunneling
  • Evaluation of existing methods for ground loss: Survey the existing formulations for ground loss, and quantitatively evaluate them based on the collected field dataset
  • Hybrid model for tunneling-induced settlement: Develop a random forest model for capturing the relationship between the ground loss and related features, which is integrated into the classical settlement solutions, yielding a hybrid model for tunneling-induced settlement

• Transfer learning for chamber pressure prediction in EPB shield tunneling
  • Transfer learning framework for chamber pressure prediction: Introduced the transfer learning methods, e.g., fine-tuning, CORAL, and MMD, for enhancing the generalization ability of existing models in predicting chamber pressure from different projects
  • Performance of the transfer learning frameworks: Utilize two filed datasets from Tianjin City and Changsha City to quantitatively demonstrate the effectiveness of the selected transfer learning methods in the few-shots problem of chamber pressure prediction
• Surrogate model for structural response under dynamical loading
  • Surrogate model development: We are the first to introduce deep neural operators e.g., DeepONet and FNO, into surrogate modeling for the response of underground structures under dynamical loading
  • Transfer learning of surrogate model: We implemented transfer learning methods for maintaining the effectiveness of the developed models under different geological conditions

Honors and Awards

• Chinese National Encouragement Scholarship - May, 2019
• First-class Prize of Chongqing Mathematical Contest in Modeling- September, 2019
• S-award of International Mathematical Contest in Modeling - March, 2020

Skills

• Programming: Python, Matlab, Opensees, LaTex
• Knowledge: Scientific machine learning, Deep learning, Soil Mechanics
• Languages: Chinese (native), English (fluent, passed IELTS with overall 7.0)