
In the development of underground space in coastal urban areas,artificial freezing is commonly used to reinforce soft soils to ensure engineering safety.However,the freeze-thaw deformation induced by the method can threaten the stability of surrounding buildings and structures.In practice,cement is used for improvement before freezing to enhance soil strength and limit deformation.However,after freeze-thaw cycles,the properties of cemented-soil degrade to some extent,and long-term loading may induce creep,leading to potential failures.Therefore,it is crucial to investigate the creep characteristics and microstructural evolution mechanisms of freeze-thaw cemented-soil.This study focuses on soft clay from Shanghai and em-ploys low field-nuclear magnetic resonance(LF-NMR)and scanning electron microscope(SEM)to investigate the influence pattern of freezing temperature and cement reinforcement on the creep characteristics of freeze-thaw cemented-soil.The re-sults indicate that,after creep,the pore volume of the unfrozen samples increases from 24.5%to 28.5%,and the particle size decreases.After freeze-thaw at-25℃,structural damage intensifies,with pore volume reaching 39.1%and particle size re-ducing to 4.50 μm,confirming that low temperature causes particle fragmentation.Freeze-thaw cycle causes a shift in the NMR T2 spectrum of samples from a single peak to a double peak.Creep facilitates the conversion of large pores to small pores,resulting in a decrease in the average pore diameter but an increase in the total number of pores,which leads to an over-all rise in porosity.This study explains the mechanism by which freeze-thaw cycle causes microstructural damage of cemented-soil and consequently changes macroscopic creep,providing a theoretical basis for the long-term stability of engi-neering projects.
Weighted signed directed graphs have attracted widespread attention for their ability to model interactions in com-plex systems via signed,directed and weighted edges.Applying Laplacian-based spectral graph convolution methods to such graphs is of great value,as these methods are both concise and interpretable.However,such applications face three major challenges:the Laplacian matrix tends to lose positive semi-definiteness,spectral decomposability and eigenvalue bounded-ness,thus limiting its scope of application;model performance is sensitive to the number of edges and weight distribution of datasets,leading to poor stability;and the computational cost is high with limited efficiency.To address these issues,this pa-per proposes a novel magnetic signed Laplacian matrix and designs a Weighted Signed Directed Graph Convolutional Net-work(WSDGCN)based on this matrix.Theoretical derivation demonstrates that this matrix preserves the core superior prop-erties of the Laplacian matrix when adapted to weighted signed directed graphs.While achieving differentiated topological characterization,it exhibits favorable robustness to edge weights.Experiments on node classification and link prediction tasks conducted on both synthetic and real-world datasets verify that the proposed method achieves superior performance compared with other competing methods.
With the increasing application of energy storage technology in microgrids,the stability of energy storage conver-ters plays a critical role in the system's performance.This paper proposes a hybrid automaton model for the Cuk topology ene-rgy storage converter,which unifies the system's discrete and continuous states.Nonlinear analysis methods are employed to quantitatively analyze the dynamic behavior of the system under variations of key parameters.In buck mode,a Thevenin equivalent circuit model for the battery is introduced,incorporating the battery's polarization effect into the system descrip-tion,to analyze its influence on the system's stability boundaries and bifurcation behavior.In boost mode,a multi-scale dy-namic model is established by introducing parametric and external excitation,and the oscillatory behavior under three typical excitation frequency ratios is analyzed using the fast-slow dynamic analysis method.The results show that as the excitation frequency ratio changes,the structure of the system's slow variables is altered,leading to the evolution of bursting oscillations from symmetric periodic bursting to asymmetric or multi-segment bursting forms.This study reveals the variation in the system's dy-namic response when parameters change,providing significant insights into the operational characteristics of the converter.
In order to comprehensively weigh the mutual constraints and design contradictions of ultra-long blades in transportation logistics,aerodynamic efficiency,structural topology,lightweight demand and service life,this paper focuses on the key technical bottlenecks and design difficulties in the research and development of large megawatt blades,system-atically discusses the structural design methods of adaptive blades,bionic blades and segmented blades,and further ex-pounds the aerodynamic shape design for power maximization and the design concept of recyclable materials for sustain-able development.The research results can provide theoretical basis and technical reference for solving the key problems of large-scale blades in development,transportation,noise reduction,operation and maintenance,improving their life cycle performance,realizing cascade utilization and green recycling,and promoting the application of new intelligent materials,bio-based materials and biodegradable materials,and have guiding significance for realizing cost reduction and efficiency increase.
In this paper,a higher-order method to solve the viscous Burgers equation is constructed and analysed.The main idea of the method is based on a proper selection of the numerical fluxes,a higher-order spatiotemporal evolution method which combines the local discontinuous Galerkin(LDG)method in the spatial direction and the spectral deferred correction(SDC)method in the temporal direction is proposed.Theoretically,the L2 stability and the optimal error estimation for the spa-tial semi-discrete LDG method are proved,and the stability of the fully discrete method is given.Numerical examples are dis-played to show the effectiveness of the proposed method.