Abstract:With the advent of the deep mining era, thermal damage in mines has become increasingly significant. The high-temperature environment in underground mines adversely impacts the physiological and psychological health of operators, reduces work efficiency, elevates the risk of accidents, and disrupts sustainable mining operations. Consequently, the development of accurate and reliable methods for classifying thermal hazards is essential for enabling mining enterprises to implement effective prevention strategies. Furthermore, such methods provide a theoretical basis for the sustainable management and utilization of geothermal energy. This study systematically considered factors influencing underground thermal damage and selected 10 quantitative indicators, encompassing both natural and human factors, as evaluation criteria. The CRITIC method was employed to determine the weight of each indicator, which was then integrated with uncertainty measurement theory to develop a novel thermal hazard assessment framework (CRITICUM). This framework enables the classification of thermal hazards in deep mine roadways. The evaluation results generated by the CRITICUM system were subsequently used to train machine learning predictive models. During the training process, the particle swarm optimization algorithm (PSO) was utilized to identify the most suitable prediction model parameters for the complex thermal environment of deep mines by leveraging its capability for continuous iterative evolution. The optimized parameters replaced the original random forest (RF) model parameters, resulting in an enhanced thermal damage prediction model (PSO-RF) with an accuracy of 96.55%, outperforming the standard RF model by 3%. Finally, the prediction model's accuracy was validated using engineering case data, demonstrating that the results met practical engineering requirements. In summary, the proposed CRITICUM-PSO-RF evaluation and prediction model can accurately classify thermal damage in deep mines and provide a valuable reference for ensuring site safety and supporting the sustainable utilization of geothermal energy.

Performance analysis of data-driven and physics-informed machine learning methods for thermal-hydraulic processes in Full-scale Emplacement experiment

Hybrid Twins Modeling of a High-Level Radioactive Waste Cell Demonstrator for Long-Term Temperature Monitoring and Forecasting

3D multi-physics uncertainty quantification using physics-based machine learning

Development of Computational Methodology in Simulating Thermal Responses of Spent Fuel in Deep Geological Disposal

Physics-informed machine learning with differentiable programming for heterogeneous underground reservoir pressure management

Machine Learning for 3D-IC Electric-Thermal Simulation and Management

Thermal Hazard Evaluation and Prediction in Deep Excavations for Sustainable Underground Mining

Long-term Thermomechanical Displacement Prediction of Energy Piles Using Machine Learning Techniques

An Innovative Computational Approach for Modeling Thermo-hydro Processes within Enhanced Geothermal System

A machine-learning-aided data recovery approach for predicting multi-material thermal behaviors in advanced test reactor capsules

Application of Machine Learning to Predict the Performance of an EMIPG Reactor Using Data from Numerical Simulations

Numerical Analysis of Thermal Process in the Near Field Around Vertical Disposal of High-Level Radioactive Waste

Probabilistic analysis of heat extraction performance in enhanced geothermal system based on a DFN-based modeling scheme

Coupled Processes at Micro- and Macroscopic Levels for Long-Term Performance Assessment Studies of Nuclear Waste Repositories

Thermo-Hydro-Mechanical Coupled Modeling of In-Situ Behavior of the Full-Scale Heating Test in the Callovo-Oxfordian Claystone

Machine Learning Modelling of Decay Heat Removal in High Temperature Gas-Cooled Reactor

Numerical Modeling on Nuclide Transport Around a Nuclear Waste Repository under Coupled Thermo-Hydro-mechanical Condition

Numerical Analysis on the Storage of Nuclear Waste in Gas-Saturated Deep Coal Seam

Physics-Informed Machine Learning Towards A Real-Time Spacecraft Thermal Simulator

Bayesian and Neural Network Approaches to Estimate Deep Temperature Distribution for Assessing a Supercritical Geothermal System: Evaluation Using a Numerical Model

Hybrid thermal modeling of additive manufacturing processes using physics-informed neural networks for temperature prediction and parameter identification