Traditional inverse neural network (INN) approaches for inverse design typically require auxiliary feedforward networks, leading to increased computational complexity and architectural dependencies. This study introduces a standalone INN methodology that eliminates the need for feedforward networks while maintaining high reconstruction accuracy. The approach integrates Principal Component Analysis (PCA) and Partial Least Squares (PLS) for optimized feature space learning, enabling the standalone INN to effectively capture bidirectional mappings between geometric parameters and mechanical properties. Validation using established numerical datasets demonstrates that the standalone INN architecture achieves reconstruction accuracy equal or better than traditional tandem approaches while completely eliminating the workload and training time required for Feedforward Neural Networks (FNN). These findings contribute to AI methodology development by proving that standalone invertible architectures can achieve comparable performance to complex hybrid systems with significantly improved computational efficiency.

Shallow-buried thick sand strata present considerable local instability risks during diaphragm wall trenching construction. However, this critical issue has not been extensively studied, despite its serious safety consequences. This paper proposes an automatic identification model for shallow-buried thick sand strata, integrating three-dimensional limit equilibrium theory with a genetic algorithm to precisely identify the most potentially dangerous local instability mass and determine its minimum safety factor. The model establishes three undetermined parameters: failure angle, upper boundary, and thickness of the local instability mass. These parameters define the search space for the local instability mass. The effectiveness of this approach was confirmed through a diaphragm wall engineering case near the Rhine River in France, where the predicted instability location closely aligned with field observations. A systematic analysis of the model indicated that the difference in slurry-groundwater levels and the friction angle are the most significant factors affecting local instability in shallow-buried thick sand strata. The model indicated that the location of the most potentially dangerous instability mass changes depending on geological conditions, and larger instability masses do not always relate to lower safety factors. Additionally, exploratory experiments revealed that support pressure losses caused by slurry infiltration significantly influence local instability calculations in sand strata. This points out the importance of considering these support pressure losses in the stability evaluations of high permeable sand strata. The results improve the evaluation of safety and the optimization of design for diaphragm wall construction in shallow-buried thick sand strata.

Underground fluid-injection operations, such as hydraulic fracturing and enhanced geothermal stimulation, have triggered multiple earthquakes across the globe. Earthquake nucleation models within the rate-and-state friction framework suggest that an increase in fluid pressure favors stable slip. However, certain observations indicate that fluid injected into faults may reduce effective normal stress, promoting fault failure, which highlights the debate on the role of fluids in controlling earthquake fault stability. This paper proposes a rate-and-state friction-based model of earthquake nucleation that incorporates fluid injection and diffusion processes, and extends the stability criteria of the system. The results show that fluid pressure heterogeneity can indeed influence fault stability. Elevated fluid pressure stabilizes faults, however, fluid pressure heterogeneity counteracts this stabilizing effect. The model suggests that pressure heterogeneity above a certain threshold facilitates seismic slip, whereas heterogeneity below this threshold can stabilize it. The results further indicate that this threshold reflects a universal instability criterion inherent to the system, rather than an incidental product of a specific fault or rock type. Accordingly, this study proposes a pressure-heterogeneity index as an operational precursor: Tracking spatiotemporal pore-pressure heterogeneity can guide the traffic-light-style adaptive control of injection. These insights provide a new, mechanism-based explanation for the role of fluids in triggering earthquakes.

The electromechanical coupling effects in two-dimensional (2D) transition metal dichalcogenides (TMDs) have attracted great interest. However, for 2D TMDs, piezoelectricity is confined to the basal plane, and the flexoelectricity-derived out-of-plane electromechanical response is usually faint, limiting the applications of this material family using the out-of-plane electromechanical effects. Here, this work reports a facile strategy to greatly enhance the out-of-plane electromechanical response of hexagonal molybdenum disulfide (2H-MoS₂) nanoflakes by stacking monolayer hexagonal boron nitride (h-BN) on 2H-MoS₂ nanoflakes to form MoS₂/BN heterostructures. The $d_{\text{33}}^{\text{eff}}$ coefficient of MoS₂/BN can reach a value comparable to that of commonly used wurtzite bulk piezoelectric materials, such as AlN and GaN. The strong out-of-plane electromechanical response of MoS₂/BN is due to the breaking of the out-of-plane structural symmetry. Kelvin probe force microscopy (KPFM) results show an increased effective work function of MoS₂/BN, indicating polar-structure formation at the heterostructure interface, which also accounts for the enhanced out-of-plane piezoresponse. This study gives an insight into the role of heterostructure engineering in the electromechanical performances of 2D TMDs, and provides this material family an opportunity for applications using out-of-plane electromechanical effects.

The perforating phase leads to complex and diverse hydraulic fracture propagation behaviors in laminated shale formations. In this paper, a 2D high-speed imaging scheme which can capture the interaction between perforating phase and natural shale bedding planes was proposed. The phase field method was used to simulate the same conditions as in the experiment for verification and hydraulic fracture propagation mechanism under the competition of perforating phase and bedding planes was discussed. The results indicate that the bedding planes appear to be no influence on fracture propagation while the perforating phase is perpendicular to the bedding planes, and the fracture propagates along the perforating phase without deflection. When the perforating phase algins with the bedding planes, the fracture initiation pressure reserves the lowest value, and no deflection occurs during fracture propagation. When the perforating phase is the angle 45°, 60° and 75° of bedding planes, the bedding planes begin to play a key role on the fracture deflection. The maximum deflection degree is reached at the perforating phase of 75°. Numerical simulation provides evidence that the existence of shale bedding planes is not exactly equivalent to anisotropy for fracture propagation and the difference of mechanical properties between different shale layers is the fundamental reason for fracture deflection. The findings help to understand the intrinsic characteristics of shale and provide a theoretical basis for the optimization design of field perforation parameters.