Search Results (504)

High-temperature structural components are susceptible to creep deformation, which can ultimately lead to failure. In this work, an AI-based framework was developed capable of predicting the creep time of 316 austenitic stainless steel. Here, creep time refers to both the time to reach specific strain levels and the time to rupture. However, the scope of the present work is limited to rupture-time prediction, while the application of the framework to strain-level prediction will be reported in future work. The dataset consisted of creep strain curves from four heats, including both rupture and non-rupture curves. Random Forest (RF), Gradient Boosting (GB), Extreme Gradient Boosting (XGB), Support Vector Regressor (SVR), Gaussian Process Regressor (GPR), and Neural Network (NN) were employed. The effects of square-root and cube-root transformations on data distribution and model learning behaviour were analysed using model learning curves. An Optuna (version 4.3.0)-based hyperparameter tuning strategy was employed. The cube-root transformation improved the learning performance of SVR, GPR, and NN, whereas RF, GB, and XGB remained unaffected. Learning curves revealed mild overfitting for RF, GB, and XGB, and very minimal overfitting for SVR, GPR, and NN. NN achieved the best predictive performance (${\mathrm{R}}^2=0.92,\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}=0.195$, deviation factor of 1.57). The findings demonstrated that the combined useof creep strain curves, data transformation, learning curve guided model selection, and rigorous hyperparameter tuning can improve the prediction accuracy under a limited dataset. Full article

Composite geomembranes (GMs) are widely used in seepage projects, where their long-term deformation properties are critical for structural safety and stability. This study conducted a 90-day creep test to compare the deformation behavior of composite GMs with varying thicknesses. Based on the long-term creep data of the composite GMs, the average values of the model parameters and the variation rules of the coefficients of variation were analyzed. When the coefficients of variation for the creep exponent a and initial strain value b were both below 10%, a short-term empirical creep model across different thickness levels was established. The accuracy and applicability of the model were analyzed by comparing its results with the measured values. The results show that the long-term deformation behavior of the composite GMs across various thicknesses aligns with a logarithmic function containing thickness-dependent coefficients. Additionally, increasing film thickness leads to reduction in the final stabilized strain. The proposed short-term model based on experimental data demonstrated reasonable agreement between the model’s 72 h fitted data and the experimental measurements. Consequently, this model may serve as a useful empirical method for predicting the long-term creep deformation of composite GMs. Full article

Carbon-fiber reinforced polymer (CFRP) tendons have attracted increasing attention as corrosion-resistant prestressing elements for prestressed concrete and cable-supported structures; however, their practical implementation requires reliable verification of long-term mechanical performance and anchorage reliability. In this study, a 9.5 mm pultruded CFRP tendon and compression-type anchorage system were developed and experimentally evaluated through relaxation, creep rupture, and fatigue tests. The tendon exhibited a tensile strength of 2501 MPa and an elastic modulus of 132.5 GPa. Relaxation tests were conducted at an initial load corresponding to 70% of the ultimate tensile capacity, and the measured relaxation loss after 1000 h was 1.02%. Based on logarithmic regression of the measured data, the relaxation loss at 1,000,000 h was estimated to be 2.11%; however, this value should be interpreted as an extrapolated long-term estimate rather than a directly verified result. Creep rupture tests performed at load ratios of 82.4–100.0% yielded an estimated 1,000,000 h creep rupture load ratio of approximately 80%, although the prediction is subject to uncertainty because of the limited number of specimens and scatter in rupture times. Fatigue tests indicated that the CFRP tendon–anchorage assembly maintained stable performance up to 2,000,000 cycles without measurable degradation in elastic stiffness under the adopted loading conditions. These results suggest that the developed CFRP tendon–anchorage system has promising potential for prestressing applications, while further long-term tests with a larger number of specimens are required to improve the statistical reliability of the extrapolated relaxation and creep rupture predictions. Full article

(1) Aluminum bonding wires are mostly used for electrical contact and transmission of electrical signals in power electronic modules. Combined cyclical mechanical and thermal loads acting on the wires can lead to premature failure of the whole module. For this purpose, based on extensive fatigue tests on a 300 µm Al-Pure wire, the authors developed, calibrated and applied a fatigue life model for a cycle range of $R=0.1$ to $R=0.7$ to other comparable aluminum wires in two previous publications. (2) Since the model is supposed to be used in an FEM post-processor for predicting the lifetime of wire bridges, the existing model was expanded in the following work. (3) Temperature dependence is included in the fatigue model, and it is made more robust in the whole possible R-range to be able to cope with the highly variable load cases in real components. In addition, a creep rupture model was developed and combined with the fatigue model by linear damage accumulation. (4) The applicability of the lifetime consumption model is demonstrated for several combined load cases. It is shown that it is necessary to consider both fatigue and creep in a combined model for a reliable lifetime prediction. Otherwise, the lifetime could be underestimated by several orders of magnitude, depending on the load case. Full article

Current landslide displacement prediction models often suffer from insufficient integration between physical mechanisms and data-driven approaches, weak model generalizability, and limited operational applicability. To address these issues, this study develops a physics-informed online learning framework for landslide displacement prediction. The core of this framework is a Physics-informed Long Short-Term Memory network (Phys-LSTM). By embedding discretized forms of the stress balance, creep constitutive, and kinematic equations as hard constraints into the LSTM’s gating mechanisms and loss function, the model ensures physically consistent predictions and enhanced interpretability throughout the learning process. Leveraging real-time data streams from the Sichuan Provincial Geological Hazard Monitoring and Warning Platform, we developed an online processing pipeline for real-time multi-source data ingestion, automated quality control, spatiotemporal alignment, and physics-informed feature engineering. A progressive three-stage learning algorithm was designed to support model cold-start, incremental training, and rolling prediction. Validation across 45 model-development landslide sites and one independent application case demonstrated the framework’s significant superiority over traditional models in displacement prediction accuracy (RMSE ≤ 1.78 mm, R² ≥ 0.96), cross-site generalization stability, and its capability to capture accelerated deformation phases. This research indicates that deeply integrating geomechanical prior knowledge into an online learning framework can effectively improve the reliability, interpretability, and operational applicability of landslide displacement prediction models, thereby providing methodological support for subsequent landslide early warning applications. Full article

The long-term deformation and stability of silty sand roadbeds subjected to repeated freeze–thaw cycles and traffic loading remain ongoing engineering concerns in seasonally frozen regions. To investigate the evolution and influencing factors of accumulative axial plastic deformation of silty sand under freeze–thaw cycles, this study focused on silty sand from a roadbed construction site in Inner Mongolia, China, a typical seasonally frozen region. Dynamic triaxial tests were conducted under loading stresses of 60–100 kPa, confining pressures of 20–60 kPa, water contents ranging from OMC to 1.2 OMC, and freeze–thaw cycles of 0–10. The results indicate that approximately 60–80% of the total accumulative axial plastic deformation occurs within the first 1000 loading cycles, after which the deformation growth rate gradually decreases. Increases in loading stress, water content, and freeze–thaw cycles promote deformation, whereas higher confining pressures suppress it. For example, increasing the confining pressure from 20 to 60 kPa reduced the final deformation from 0.16% to 0.07%, while increasing the number of freeze–thaw cycles from 0 to 10 increased the final deformation from 0.10% to 0.28%. Based on the experimental data, a new predictive model considering net stress, octahedral shear stress, water content ratio, and freeze–thaw cycles was developed. The model demonstrates high accuracy in predicting accumulative plastic deformation, with a coefficient of determination of 0.915, and is applicable to both plastically stable and weakly plastic creep conditions. This study provides a reference for the design, construction, and mitigation of subgrade damage in silty sand roadbeds in seasonally frozen regions. Full article

Accurate prediction of slope displacement from high-frequency GNSS monitoring data is critical for early warning of landslides and tailings dam failures. However, existing deep learning approaches often neglect the spatial coordination imposed by geological structures and fail to decouple abrupt deformation signals from background noise, leading to non-physical oscillations and inconsistent long-term predictions. To address these limitations, this paper proposes a Symmetry-Aware Physics-Guided Spatio-Temporal Graph Network (PG-STGN). First, a geological hierarchy-aware graph is constructed by integrating geometric proximity with prior knowledge of exploration levels, where the resulting adjacency matrix is symmetric by design and reflects the physical symmetry of deformation interactions among monitoring points at the same elevation. A hierarchical masking mechanism restricts feature aggregation to physically connected neighborhoods while preserving this symmetry. Second, an improved dual-path temporal convolutional network (iTCN) decouples high-frequency abrupt variations from low-frequency evolutionary trends, enabling both sensitive detection of sudden deformation and stable tracking of long-term creep. Third, a physics-consistent loss function combining first-order temporal differencing and graph Laplacian regularization enforces kinematic smoothness and spatial coordination; the Laplacian itself is derived from the symmetric adjacency matrix, ensuring symmetric regularization across the monitoring network. Evaluated on a real-world slope GNSS dataset from a large-scale mining project, PG-STGN reduces mean squared error (MSE) by approximately 23.7% and achieves a global R² of 0.924, outperforming state-of-the-art spatio-temporal models. Ablation studies confirm that the symmetric physics-guided graph, dual-path decoupling, and consistency loss are each essential for suppressing spurious correlations and maintaining physically plausible predictions. The proposed framework provides a robust, interpretable, and symmetry-constrained solution for automated slope monitoring under complex geological conditions. Full article

Polyethylene (PE) pipes are widely employed in urban gas and water conveyance systems due to their excellent corrosion resistance, cost efficiency, and long service life. However, creep-induced delayed failure remains a critical threat to long-term operational safety and may lead to leakage accidents. Accurate and efficient prediction of creep rupture life is essential for risk control and structural design. This study investigated the performance of four commercial polyethylene pipes, including two PE80-grade and two PE100-grade pipes. By combining the creep test with the critical strain criterion, an efficient and reliable method for predicting the ductile failure lifetime was developed. Creep tests were carried out on dumbbell specimens cut from PE pipes under multiple temperature and stress levels. The time-hardening model was adopted to characterize the nonlinear viscoelastic creep evolution, and the ductile failure time was determined by introducing the critical strain threshold. The predicted lifetimes were systematically validated against experimental data from long-term hydrostatic tests. Results show that the predicted failure times agree well with the measured values, verifying the accuracy and engineering applicability of the proposed method. This approach provides a high-efficiency alternative to conventional long-term hydrostatic tests, offering valuable support for material selection, safety evaluation, and engineering design of PE pipeline systems. Full article

Lightweight alloys in aerospace precision structures undergo slow but cumulative creep deformation during long-term storage, wherein strain accumulation over years can compromise dimensional stability and operational reliability. However, continuum damage mechanics (CDM) constitutive models, while physically grounded, require extensive parameter calibration and exhibit degraded accuracy during the primary creep stage. Meanwhile, purely data-driven approaches are impractical for the sparse datasets typical of accelerated creep testing, wherein as few as 14 data points may be available per condition. Although physics-informed neural networks (PINNs) have shown promise in computational mechanics, existing PINN-based creep studies predict only scalar life quantities rather than the full strain–time curve $\epsilon \left(t\right)$, and none embed damage evolution equations as differential constraints. This study proposes a damage-coupled PINN framework (termed DC-PINN) that predicts the complete creep strain evolution $\epsilon \left(t\right)$ by embedding CDM damage evolution ordinary differential equations (ODEs) as hierarchical differential constraints within the learning process. The framework couples the predicted strain rate $\mathrm{d}\epsilon /\mathrm{d}t$ with the damage state $D\left(t\right)$ through material-specific constitutive ODEs, supplemented by monotonicity enforcement and boundary conditions. Alloy-specific formulations are developed for 2A12-T4 aluminum (Arrhenius kinetics, no damage) and ZM6 magnesium (Sandström dislocation model with Ostwald-ripening-driven grain coarsening damage). Validated on 13 experimental conditions spanning both alloys (50–100 °C, 20–60 MPa, 14–100 points per condition), DC-PINN achieves $R^2>0.99$ for 2A12-T4 and $R^2>0.97$ for ZM6 across all tested conditions. Ablation studies show that the total physics-driven $R^2$ improvement is 5.8 times larger for the data-sparse ZM6 (14–34 points) than for the data-rich 2A12-T4 (∼100 points), with the CDM damage coupling alone accounting for 22% of the improvement in ZM6. To the best of our knowledge, this represents the first integration of CDM damage evolution ODEs as differential constraints within PINNs for creep strain modeling, providing a physically consistent and data-efficient tool for the storage life assessment of aerospace structures. Full article

The macroscopic resistance of asphalt mixtures to permanent deformation is fundamentally governed by the mechanical properties of the constituent asphalt mortar; however, a unified evaluation system that quantitatively links the energy evolution between these two scales is currently lacking. This study aims to bridge this gap by establishing a multiscale framework to characterize and predict the recoverable and dissipated energy behaviors of asphalt materials. To achieve this, Multi-Stress Creep Recovery (MSCR) tests and Multi-Sequence Repeated Loading (MSRL) tests were conducted on asphalt mortar and mixtures, respectively, to capture energy evolution under varying stress, temperature, and gradation conditions. Subsequently, Multiple Linear Regression (MLR) and Artificial Neural Network (ANN) models were developed to correlate mesoscopic mortar parameters with macroscopic mixture performance. Experimental results reveal that energy indicators are significantly influenced by loading stress and aggregate skeleton, with finer gradations exhibiting greater responsiveness to stress changes. A strong cross-scale dependency was identified, evidenced by a correlation coefficient of 0.86 between the recoverable energy of the mixture ($U_{r-mix}$) and that of the mortar ($U_{r-mortar}$). Furthermore, the developed ANN model demonstrated exceptional predictive accuracy ($R^2\ge 0.99$) in upscaling energy indicators. This study develops a multiscale energy framework that integrates experimentally derived energy indicators from asphalt mortar and asphalt mixture, enabling the prediction of macroscopic mixture performance from mesoscopic mortar energy evolution rather than relying solely on empirical machine-learning correlations. Full article

Fluid flow in fractured-vuggy carbonate reservoirs is characterized by extreme multiscale heterogeneity, where the coexistence of tight matrix rock and macroscopic cave challenges traditional Darcy-based continuum models. This paper presents a semi-analytical solution for two-phase immiscible displacement in a one-dimensional composite porous–cavernous–porous (PCP) system. The main feature of the model is that the cave region is treated separately from the porous domains: classical Darcy flow is used in the surrounding matrix, whereas an idealized free-flow representation is introduced for open caves based on a simplified one-dimensional treatment of the cave momentum balance. To elucidate the impact of distinct flow regimes on displacement dynamics, three physical models are compared for the cave region: (1) an open-cave model represented by a simplified free-flow formulation; (2) a filled-cave non-Darcy model governed by the Forchheimer equation using the Ergun correlation; and (3) a creeping-flow model governed by Darcy’s law. A piecewise semi-analytical solution procedure is established to enforce flux continuity, characterize interfacial state remapping, and determine the downstream front under global water-balance closure. The results show that both cave geometry and internal cave-flow mechanism critically control water-front advancement. While the open-cave model exhibits piston-like displacement behavior with high local displacement efficiency but stronger preferential flow, the Forchheimer model shows that inertial resistance can modify the saturation profile and delay breakthrough relative to the Darcy prediction. The proposed framework provides an idealized theoretical reference for benchmarking numerical simulators and for interpreting waterflooding behavior in complex vuggy reservoirs under one-dimensional, incompressible, gravity-free, and capillarity-free conditions. Full article

Nickel-based superalloys are key materials for aerospace and gas turbine applications. Traditional manufacturing approaches struggle to produce controllable porous structures with complex topologies. This review focuses on additively manufactured porous Ni-based superalloys, and summarizes progress in porous structure design, including disordered, lattice, TPMS, bio-inspired, and AI-assisted structures. Common additive manufacturing technologies are introduced, along with their effects on microstructure evolution and defect formation. The review discusses non-equilibrium microstructures, elemental segregation, and typical defects such as lack-of-fusion, keyhole porosity, and residual stress, as well as their influences on strength, fatigue, and creep behavior. Post-processing strategies for defect mitigation and performance optimization are also summarized. This review highlights the unique mechanical and physical behavior of porous structures compared to bulk materials, with an emphasis on anisotropy, stress localization, and defect sensitivity. Finally, several critical and specific challenges are identified, including multi-scale modeling, microstructure control in complex topologies, fatigue prediction, and physics-constrained AI design. This review aims to provide a clear, focused, and structurally consistent overview of the current state of the field, and to support future research on additively manufactured porous Ni-based superalloys. Full article

Maintaining the integrity of impermeable strata between mine workings and overlying aquifers is critical, because seepage pathways may cause mine flooding and surface subsidence. In the Upper Kama potash deposit, the impermeable sequence is a 50–140 m thick layered sequence of evaporites and clays overlying mined-out chambers. Under long-term loading, salt rocks tend to creep, soften, and localize damage, which can cause failure in the impermeable strata. In this paper, the Concrete damage-plasticity model, supplemented by the N2PC-MCT viscoplastic creep model, is applied to simulate the initiation and evolution of seepage pathways in the Upper Kama impermeable strata. Model parameters are obtained from published laboratory tests (uniaxial and triaxial compression and tension) and validated using observed ground-surface subsidence. A plane-strain finite-element model incorporates the stratified lithology, interface elements between layers, and sequential excavation. Long-term simulations up to 50 years investigate two operational scenarios: with and without backfilling. The calibrated model reproduces the main stages of surface subsidence and chamber closure. Without backfilling, simulations indicate that tensile damage localizes mainly in a stiff central salt layer of the impermeable strata, with most cracks appearing approximately between 33 and 37 years after the start of mining. With backfill, tensile crack propagation stops and damage remains stable. A hypothetical homogeneous impermeable strata case confirms that the observed central-layer cracking is associated with stiffness contrasts and composite bending in the stratified system. An approximate analytical multilayer beam solution, based on energy minimization, predicts bending stress concentration in stiff intermediate layers and is consistent with the numerical stress distribution. The combined numerical and analytical results provide insight into the mechanisms of long-term conductive fracture initiation in stratified impermeable strata and may serve as a basis for preliminary hazard indication and for planning mitigation measures, including backfilling and focused monitoring of stiff central layers. Because the study is based on a 2D plane-strain model, the quantitative estimates should be regarded as preliminary and require verification by 3D modelling and further field observations. Full article

In order to investigate the influence of notched structures on creep performance under long-term high-temperature conditions, durability tests were conducted on ring-notched and hole-containing thin tubular specimens of directional columnar-grain DZ411 alloy at 850 °C and 930 °C. The results were compared with those of smooth round rod specimens at same temperatures and stress levels, to evaluate the impact of notched structures described above on the rupture life. Based on the experimental data, a finite element subroutine was developed using a macroscopic phenomenological creep model to simulate the creep deformation behavior of the structural components. The stress relaxation characteristics of the two types of notched structures were analyzed. The results show that the ring-notched structure exhibits significant stress relaxation, leading to a “notch strengthening” effect, which improves the endurance property; conversely, the small-hole structure shows insufficient stress relaxation, resulting in “notch weakening” and a reduction in the endurance property. The developed subroutine demonstrates sufficient engineering accuracy in notch creep simulation. Using creep strain as the fracture criterion, the predicted endurance life showed a deviation from experimental results within the acceptable engineering range, indicating that the subroutine has sufficient engineering accuracy. Full article

The accurate prediction of creep deformation is essential for the reliable use of stainless steel 316 in high-temperature applications. Conventional creep models employ fixed material parameters and often fail to capture the evolving deformation mechanisms that are active during long-term service. In this work, a novel physics-guided creep model is proposed, incorporating adaptive stress sensitivity and dynamic activation energy terms optimized using machine learning techniques. The model is calibrated using extensive experimental creep data and compared with classical analytical models and purely data-driven approaches. The results show that the proposed hybrid framework significantly improves predictive accuracy across all creep stages while retaining physical interpretability. Full article