Danhua Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Danhua Zhu has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Tianlong Su
School of Resources, Environment and Materials, Guangxi University, Nanning 530004, China
2 shared publications - Xiaoquan Li
School of Resources Environment and Materials, Guangxi University, 100 University East Road, Nanning 530004, China
2 shared publications - Xuefeng Li
School of Resources, Environment and Materials, Guangxi University, Nanning 530004, China
2 shared publications - Xuzhao Yuan
School of Resources, Environment and Materials, Guangxi University, Nanning 530004, China
2 shared publications - Han Du
School of Resources, Environment and Materials, Guangxi University, Nanning 530004, China
1 shared publication - Linxin Zhang
Guangxi Industrial Design Group Co., Ltd., Nanning 530023, China
1 shared publication
Latest Publications
- Open Access
PS-InSAR Monitoring Integrated with a Bayesian-Optimized CNN–LSTM for Predicting Surface Subsidence in Complex Mining Goafs Under a Symmetry Perspective
Mine-induced surface subsidence threatens infrastructure and can trigger cascading geohazards, so accurate and computationally efficient monitoring and forecasting are essential for early warning. We integrate Persistent Scatterer InSAR (PS-InSAR) time series with a Bayesian-optimized CNN–LSTM designed for spatiotemporal prediction. The CNN extracts spatial deformation patterns, the LSTM models temporal dependence, and Bayesian optimization selects the architecture, training hyperparameters, and the most informative exogenous drivers. Groundwater level and backfilling intensity are encoded as multichannel inputs. Endpoint anchoring with affine calibration aligns the historical series and the forward projections. PS-InSAR indicates a maximum subsidence rate of 85.6 mm yr−1, and the estimates are corroborated against nearby leveling benchmarks and FLAC3D simulations. Cross-site comparisons show acceleration followed by deceleration after backfilling and groundwater recovery, which is consistent with geological engineering conditions. A symmetry-aware preprocessing step exploits axial regularities of the deformation field through mirroring augmentation and documents symmetry-breaking hotspots linked to geological heterogeneity. These choices improve generalization to shifted and oscillatory patterns in both the spatial CNN and the temporal LSTM branches. Short-term forecasts from the BO–CNN–LSTM indicate subsequent stabilization with localized rebound, highlighting its practical value for operational planning and risk mitigation. The framework combines automated hyperparameter search with physically consistent objectives, reduces manual tuning, enhances reproducibility and generalizability, and provides a transferable quantitative workflow for forecasting mine-induced deformation in complex goaf systems.
Open AccessPS-InSAR Monitoring Integrated with a Bayesian-Optimized CNN–LSTM for Predicting Surface Subsidence in Complex Mining Goafs Under a Symmetry Perspective
Mine-induced surface subsidence threatens infrastructure and can trigger cascading geohazards, so accurate and computationally efficient monitoring and forecasting are essential for early warning. We integrate Persistent Scatterer InSAR (PS-InSAR) time series with a Bayesian-optimized CNN–LSTM designed for spatiotemporal prediction. The CNN extracts spatial deformation patterns, the LSTM models temporal dependence, and Bayesian optimization selects the architecture, training hyperparameters, and the most informative exogenous drivers. Groundwater level and backfilling intensity are encoded as multichannel inputs. Endpoint anchoring with affine calibration aligns the historical series and the forward projections. PS-InSAR indicates a maximum subsidence rate of 85.6 mm yr−1, and the estimates are corroborated against nearby leveling benchmarks and FLAC3D simulations. Cross-site comparisons show acceleration followed by deceleration after backfilling and groundwater recovery, which is consistent with geological engineering conditions. A symmetry-aware preprocessing step exploits axial regularities of the deformation field through mirroring augmentation and documents symmetry-breaking hotspots linked to geological heterogeneity. These choices improve generalization to shifted and oscillatory patterns in both the spatial CNN and the temporal LSTM branches. Short-term forecasts from the BO–CNN–LSTM indicate subsequent stabilization with localized rebound, highlighting its practical value for operational planning and risk mitigation. The framework combines automated hyperparameter search with physically consistent objectives, reduces manual tuning, enhances reproducibility and generalizability, and provides a transferable quantitative workflow for forecasting mine-induced deformation in complex goaf systems.
- Open Access
Multi-Factor Analysis and Graded Remediation Strategy for Goaf Stability in Underground Metal Mines: Fluid–Solid Coupling Simulation and Genetic Algorithm-Based Optimization Approach
To ensure the green, safe, and efficient extraction of mineral resources and promote sustainability, the stability of mined-out areas has become a critical factor affecting safe production and ecological restoration in underground metal mines. The instability of underground goafs poses a significant threat to mine safety, especially when irregular excavation patterns interact with high ground stress, exacerbating instability risks. Most existing studies lack a systematic and multidisciplinary integrated framework for comprehensive evaluation and management. This paper proposes a trinity research system of “assessment–optimization–governance”, integrating theoretical analysis, three-dimensional fluid–solid coupling numerical simulation, and a filling sequence optimization method based on genetic algorithms. An analysis of data measured from 243 pillars and 49 goafs indicates that approximately 20–30% of the pillars have a factor of safety (FoS) below 1.0, signaling immediate instability risks; additionally, 58% do not meet the threshold for long-term stability (FoS ≥ 1.5). Statistical and spatial analyses highlight that pillar width-to-height ratio (W/H) and cross-sectional area significantly influence stability; when W/H exceeds 1.5, FoS typically surpasses 2.0. Numerical simulations reveal pore water pressures of 1.4–1.8 MPa in deeper goafs, substantially reducing effective stress and accelerating plastic zone expansion. Stability classification categorizes the 49 goafs into 7 “poor”, 37 “moderate”, and 5 “good” zones. A genetic algorithm-optimized filling sequence prioritizes high-risk area remediation, reducing maximum principal stress by 60.96% and pore pressure by 28.6%. Cemented waste rock filling applied in high-risk areas, complemented by general waste rock filling in moderate-risk areas, significantly enhances overall stability. This integrated method provides a scientific foundation for stability assessment and dynamic remediation planning under complex hydrogeological conditions, offering a risk-informed and scenario-specific application of existing tools that improves engineering applicability.
Open AccessMulti-Factor Analysis and Graded Remediation Strategy for Goaf Stability in Underground Metal Mines: Fluid–Solid Coupling Simulation and Genetic Algorithm-Based Optimization Approach
To ensure the green, safe, and efficient extraction of mineral resources and promote sustainability, the stability of mined-out areas has become a critical factor affecting safe production and ecological restoration in underground metal mines. The instability of underground goafs poses a significant threat to mine safety, especially when irregular excavation patterns interact with high ground stress, exacerbating instability risks. Most existing studies lack a systematic and multidisciplinary integrated framework for comprehensive evaluation and management. This paper proposes a trinity research system of “assessment–optimization–governance”, integrating theoretical analysis, three-dimensional fluid–solid coupling numerical simulation, and a filling sequence optimization method based on genetic algorithms. An analysis of data measured from 243 pillars and 49 goafs indicates that approximately 20–30% of the pillars have a factor of safety (FoS) below 1.0, signaling immediate instability risks; additionally, 58% do not meet the threshold for long-term stability (FoS ≥ 1.5). Statistical and spatial analyses highlight that pillar width-to-height ratio (W/H) and cross-sectional area significantly influence stability; when W/H exceeds 1.5, FoS typically surpasses 2.0. Numerical simulations reveal pore water pressures of 1.4–1.8 MPa in deeper goafs, substantially reducing effective stress and accelerating plastic zone expansion. Stability classification categorizes the 49 goafs into 7 “poor”, 37 “moderate”, and 5 “good” zones. A genetic algorithm-optimized filling sequence prioritizes high-risk area remediation, reducing maximum principal stress by 60.96% and pore pressure by 28.6%. Cemented waste rock filling applied in high-risk areas, complemented by general waste rock filling in moderate-risk areas, significantly enhances overall stability. This integrated method provides a scientific foundation for stability assessment and dynamic remediation planning under complex hydrogeological conditions, offering a risk-informed and scenario-specific application of existing tools that improves engineering applicability.