SHANGTONG YANGROCK MECHANICS · SUBSURFACE ENERGY

ROCK MECHANICS · SUBSURFACE ENERGY

Shangtong Yang

Professor

Professor, China University of Mining and Technology, Xuzhou, China

Reader, University of Strathclyde, Glasgow, UK

We research subsurface energy flow through natural and engineered rock systems

Professor Shangtong Yang

Academic Service

Academic committees

Editorial service

01

Research

01

Rock fatigue and fracture

In-situ XCT DVC analysis

We investigate how internal microcracking and local deformation govern the macroscopic failure and hysteresis of crystalline rock. In-situ X-ray computed tomography (XCT), quantitative void segmentation and digital volume correlation (DVC) connect evolving three-dimensional crack networks with internal strain fields. Through uniaxial compression of fine granite, we have revealed the nucleation of isolated microcracks followed by unstable coalescence, with damage evolving from predominantly inclined cracking towards a stronger contribution from load-parallel cracks near failure [paper]. Moreover, a complete cycle of Loading–unloading–reloading observations link the mechanical hysteresis to delayed strain release in dilation zones associated with shear cracking, and reveal additional damage before the previous peak load is regained [paper]. These observations underpin micromechanical models of damage accumulation and rock fatigue.

Fatigue constitutive modelling

We have developed constitutive descriptions that connect cyclic loading histories to progressive damage and fracture in rocks and other quasi-brittle materials. An S–N-based fatigue degradation law was first coupled with a fluid cohesive crack model for hydraulic rock fracture [paper]. The subsequent nonlinear cyclic cohesive zone model distinguishes fatigue driven irreversible displacement from static damage induced stiffness loss, with fatigue evolution governed by the current stress and damage state [paper]. Validation against concrete three-point bending tests reproduces crack initiation, stable growth and rapid fracture, and captures loading sequence effects that linear cumulative damage rules cannot represent. Together, these developments can provide a framework for modelling fatigue-controlled fracture across quasi-brittle materials.

02

Hydraulic stimulation technologies

Fatigue-inspired hydraulic pulse stimulation

We explore how controlled hydraulic pulses can initiate rock fractures at lower pressure while managing subsequent crack growth. A finite element framework couples S–N based fatigue degradation, cohesive fracture, fluid flow and pore pressure, validated against laboratory pulsating fracture results [paper]. In the modelled cases, pulses reduce breakdown pressure by 10–18% over 10–4,000 cycles of applied pressure. The simulations also show that pressure controlled pulses can drive rapid crack growth after initiation. A three stage strategy is proposed to switch from high frequency pressure pulses to flow controlled cyclic injection, then lowers frequency while increasing the flow as the fracture extends. The resulting simulations demonstrate slower and steadier propagation and identify how injection control can exploit fatigue without losing control of fracture growth.

Near-wellbore fracture control and natural fracture activation

We investigate how well construction, perforations and existing fractures shape hydraulic stimulation near the wellbore. A coupled analytical–numerical model represents pressure transmission through the casing–cement–rock system, while a hydro-mechanical mixed-mode cohesive model incorporates frictional slip on natural fractures [paper]. The studies show that larger pressure transmission reduces near-wellbore fracture deflection but raises breakdown pressure. Hydraulic–natural fracture interactions reproduce crossing, deflection and offsetting; deflection is shear-dominated, and the outcome depends on approach angle, fracture strength, in-situ stress and injection conditions. In the simulated near-wellbore configurations, low injection rates can favour the natural fracture activation, providing a mechanistic basis for well design and stimulation planning [paper].

03

Subsurface energy storage systems

Lined rock caverns energy storage

We assess the integrity of lined rock caverns by resolving interactions between the steel barrier, concrete lining and fractured host rock. A phase-field framework combines mixed-mode cracking, frictional contact, steel plasticity, discrete fracture networks and spatial material variability [paper]. The simulations reveal a design trade-off: stronger concrete improves overall confinement but can concentrate cracking and transfer local strain into the steel lining. Fracture connectivity controls how damage spreads into the rock, while greater burial depth increases confinement and suppresses crack opening. Pressure–depth maps quantify these competing effects and support site specific operating limits.

Thermal energy storage

We connect the energy performance of subsurface thermal stores with the durability of their containment materials. Lifetime thermo-economic modelling couples flooded mineshaft water, surrounding ground and district-heating demand: in the cost-optimal scenarios studied, daily-to-weekly operation keeps heat losses below 10% of input energy, whereas seasonal storage incurs much larger losses [paper]. Complementary thermo-chemical experiments investigate mortar lining under 15–55 °C cycling and 3 wt.% sulphate exposure. The combined conditions produce severe accelerated deterioration, with leaching and ettringite formation creating depth dependent stiffness and thermal deformation that promote cracking during cooling [paper]. These studies show why operational optimisation and material degradation must be evaluated together when planning longer service life thermal storage.

04

Whole life-cycle management and AI application

Environmental deterioration and performance evolution

We develop mechanistic models linking environmental deterioration to cracking and serviceability loss in reinforced concrete structures. A von Mises-based corrosion model developed describes the non-uniform growth of rust around reinforcement and supplies expansion to mesoscale fracture models containing aggregates, mortar and interfacial transition zones [paper]. The simulations capture microcrack initiation at interfaces, coalescence into surface cracks, and toughening through crack deflection, aggregate bridging and microcrack shielding. Mixed-mode analysis shows that a uniform-corrosion assumption can delay predicted surface cracking and underestimate crack width, particularly for thicker covers [paper]. These results can provide a basis for interpreting damage evolution and assessing the durability of concrete infrastructure and containment systems.

Machine-learning inspired analysis

We combine physics based fracture simulations with machine learning for parameter identification and time-dependent reliability assessment. An artificial neural network trained on mesoscale tensile simulations inversely identifies concrete interface strength and fracture energy from measured stress–displacement curves, reproducing the RILEM benchmark across different aggregate arrangements [paper]. A complementary surrogate-assisted reliability framework links non-uniform corrosion, crack width predictions and Monte Carlo sampling to estimate serviceability failure over time [paper]. In the worked example, accounting for non-uniform corrosion shortens the remaining safe life after corrosion initiation by about 32% relative to a conventional uniform model. These approaches translate detailed mechanics into efficient tools for material calibration and maintenance decisions, within the validated modelling conditions.

Recent journal publications

* Corresponding author

2026

2025

2024

Team

Postdoctoral researcher

Dr Dongsheng Chen

Dr Dongsheng Chen

Postdoctoral researcher

Rock physics and mechanics

PhD students

Zhiming Feng

Zhiming Feng

PhD student

Rock fatigue and hydraulic pulse fracturing

Jucan Yao

Jucan Yao

PhD student

Whole life-cycle assessment of LRC CAES

Kangda Chen

Kangda Chen

PhD student

Micro-mechanism of rock fatigue

Tianran Lin

Tianran Lin

PhD student

New lining systems for LRC hydrogen storage

Bin Yao

Bin Yao

PhD student

Pulse generation and application in subsurface engineering

Bo Yang

Bo Yang

PhD student

Numerical simulation and framework for heat storage in legacy mine space

Ruiyu He

Ruiyu He

PhD student

Geo-grouting and Geochemistry

Master’s students

Yuhui Gong

Yuhui Gong

Master's student

Experimental leakage study on Oil/Gas well integrity

Yunfei Yang

Yunfei Yang

Master's student

Molecular dynamics study on rock shearing

Tianle Zhu

Tianle Zhu

Master's student

Ion transport in linings of LRC energy storage

Yilei Kong

Yilei Kong

Master's student

TBD

Lujia Li

Lujia Li

Master's student

TBD

Zhenyu Wu

Zhenyu Wu

Master's student

TBD

Past group members

  1. Rocco Panetta, 2022-2026 (now Civil Engineer at Transport Scotland)
  2. Huachuan Wang, 2022-2025 (now Professor at Shandong University)
  3. Qinxin Hu, 2020-2025 (now Research Fellow at Lawrence Berkeley National Laboratory / UC Santa Cruz)
  4. Feng Zhang, 2020-2025 (now Postdoctoral researcher at Hunan University)
  5. Yike Dang, 2023-2024 (now Assistant Professor at Xi'an University of Technology)
  6. Huicong Xu, 2022-2023 (now Professor at Xi'an University of Science and Technology)
  7. Reza Ghanbari, 2022-2023 (now Senior Engineer at Arup)
  8. Xun Xi, 2017-2022 (now Professor at University of Science and Technology Beijing)
04

Vacancy

Research opportunities

We currently accept applications for research positions in our group, including Research Associate, Research Assistant, PhD and research masters. If you are interested in joining us, please feel free to contact the team manager at tb26220031a41@cumt.edu.cn

Contact the team manager
05

Contact & collaboration

Team@CUMT, Xuzhou, China

shangtong.yang@cumt.edu.cn