Applied Composite Materials最新文献

Machine learning (ML) is revolutionizing the development and optimization of composite materials by enabling data-driven approaches for material design, manufacturing processes, and performance prediction. Widely applied in aerospace, automotive, defense, and biomedical engineering, composite materials require precise engineering to achieve targeted mechanical, thermal, and structural properties. Traditional methods, which rely heavily on extensive experimentation and expert knowledge, are often costly, time-consuming, and limited in simulating extreme conditions. ML techniques—including supervised, unsupervised, semi-supervised, and reinforcement learning—allow efficient analysis of complex datasets, pattern recognition, and accurate prediction of material behavior. This review summarizes recent advances in ML applications for composites, covering material design, process optimization, multiscale property prediction, and structural health monitoring. Case studies demonstrate significant improvements in property estimation, defect reduction, and manufacturing efficiency. Challenges such as high-dimensional data, overfitting, and model interpretability are discussed. Addressing these issues is critical for enabling AI-driven development of next-generation composites with enhanced performance, reliability, and sustainability.

Transverse stabilizer bars are key components of automotive suspension systems, primarily enhancing vehicle roll resistance and driving stability. However, conventional steel stabilizer bars suffer from excessive weight and limited fatigue durability. To overcome these limitations, this study proposes a lightweight design for composite transverse stabilizer bars based on multi-fiber hybrid-reinforced composites, for which a stiffness prediction theory and a finite element analysis (FEA) model were established. Experimental validation confirmed the model’s reliability, with prediction errors below 5%. The results indicate that triaxial braided structures significantly improve the stiffness and fatigue life of the composite stabilizer bars. Among the various configurations, a hybrid triaxial braided structure combining a glass-fiber outer layer with a ramie-fiber inner layer exhibited the best overall performance. Compared with steel transverse stabilizer bars of identical dimensions, the proposed composite design achieves a 52.2% reduction in weight while maintaining equivalent stiffness, along with a 6.07% improvement in fatigue life. This research provides both theoretical and experimental foundations for the engineering application of composite stabilizer bars, which hold significant potential for advancing automotive lightweighting and sustainable manufacturing.

Delamination constitutes a critical failure mechanism in composite structures under impact conditions, significantly compromising structural integrity through progressive interfacial energy dissipation. An understanding of dynamic delamination and fracture behavior of UHMWPE laminates is developed in this study, with focus on the influence of loading rates and specimen configurations. A specialized test apparatus combined with optimized wedge-insert double cantilever beam specimens and high-speed optical imaging is formulated to characterize deformation and crack propagation. Results show that UHMWPE laminates exhibit significant nonlinearity during fracture, attributed to the inherent ductility and plastic deformation, distinct from brittle composites. Specimen thickness and interlaminar ply orientation considerably affect fracture responses. The 90°/90° interfacial configuration effectively suppressed fiber bridging and crack migration, while increased thickness reduces plastic artifacts, ensuring measurement of intrinsic toughness and reproducibility. The fracture toughness demonstrates strong loading-rate dependence, with noticeable enhancement at high loading rates. A rate-dependent cohesive zone model is developed and validated. The close correspondence between numerical and experimental results confirms the quantitative accuracy and mechanistic fidelity. The findings provide experimental references and theoretical support for determining delamination behavior of ballistic composites, with direct relevance to performance evaluation and lightweight design.

Fiber-reinforced polymer (FRP) composites are widely used in marine and offshore structures, yet long-term exposure to seawater degrades their mechanical response and alters failure mechanisms. While these effects are well documented experimentally, they are less frequently propagated into finite element (FE) material models used for structural prediction. This study examines how artificial seawater aging modifies the parameter set of LS-DYNA’s MAT54 (Enhanced Composite Damage) model. Mechanical data (tension, compression, and in-plane shear) were compiled for four laminate systems (glass/epoxy, glass/vinyl ester, glass/polyester, and carbon/epoxy) in both unaged and aged states. Stiffness, strength, and damage/softening parameters were investigated using calibration against the measured stress–strain behavior. Aging was represented through targeted updates to MAT54 inputs rather than ad hoc knock-down factors, enabling the model to capture coupled reductions in stiffness and strength across loading modes. The outcome is a set of aging-aware MAT54 parameter tables for each composite system that provide consistent fits to experimental behavior and a practical basis for FE simulation of environmentally exposed laminates. The analysis also indicates which (matrix-dominated) parameters are most sensitive to seawater exposure, offering guidance for selecting aging-dependent inputs. The calibrated datasets and workflow support more reliable prediction of structural performance and damage progression in FRP composites subjected to marine environments.

Traditional cellulose-based paper suffers from inherent flammability, hydrophilicity, and oxidative degradation. These limitations cause irreversible damage, including yellowing, moisture absorption, and structural decay, during long-term archival storage. To address these challenges, a biomimetic strategy inspired by the hierarchical architecture of bird-nest (interwoven branches, grass, and mud) is proposed to engineer a multifunctional composite paper. By integrating aluminum silicate fibers (ASF) as load-bearing “branches”, ultralong hydroxyapatite nanowires (HAPNW) as flexible “grass”, and an inorganic adhesive(aluminum oxide sol) as binding “mud”, a nest-like network is constructed to synergistically enhance mechanical robustness, flame resistance, and environmental stability. Unlike pure HAPNW paper with limited tensile strength (25 MPa), the optimized composite (30% ASF) achieves a remarkable strength of 62 MPa and modulus of 9 GPa, attributed to stress redistribution via interfacial fiber entanglement and inorganic binder-mediated cohesion. Further hydrophobic modification confers superhydrophobicity and self-cleaning capabilities, while retaining writeability comparable to conventional paper. Crucially, the composite exhibits non-combustibility under direct flame exposure (700°C, 60 s) and exceptional thermal stability (84% residual mass at 800°C), outperforming commercial counterparts. This work pioneers a biomimetic paradigm for designing high-performance archival materials, offering a sustainable solution to reconcile mechanical durability, fire safety, and environmental resilience in paper-based preservation technologies.

Carbon fibre-reinforced thermoplastics used in structural applications, such as crash-relevant parts, are typically subjected to different strain rate conditions. This requires a deep understanding of their structural behaviour under these conditions. To date, there is limited data on their strain rate dependent behaviour. In this study, four-point flexure tests are used to determine the longitudinal flexural properties and failure behaviour of unidirectional carbon fibre-reinforced polyamide 6 (CF-PA6) material as a function of the applied strain rate. A universal testing machine and a drop tower test setup with two different drop heights were used to investigate the behaviour in quasi-static (1.3 × 10^− 4 1/s) to medium strain rates (7 1/s). Flexural modulus, flexural strength and fracture strain were determined for all strain rates applied, and the fracture morphology was analysed using microscopic images and high-speed camera recordings. Further, Weibull analysis was performed on the flexural strength. The experimental investigation of the flexural properties showed no significant strain rate effect and no trend was observed in the Weibull parameters. Irrespective of the strain rate applied, all specimens tested exhibited a comparable fracture morphology due to a similar failure process of initial compressive failure followed by tensile failure.

SiCp/Al composites, with high specific strength, low thermal expansion coefficient, and excellent wear resistance, are widely used in aerospace lightweight components. However, during machining, hard and brittle SiC particles not only accelerate tool wear but also restrict low-damage machining efficiency due to complex chip formation mechanisms and uncontrollable surface damage. Thus, focusing on low-volume-fraction SiCp/Al, this study investigates the chip formation mechanism and its impact on surface morphology, revealing how cutting parameters regulate chip morphology and surface quality to provide theoretical support for process optimization. A three-phase SiCp/Al thermo-mechanical coupling model was built, combined with 2D cutting simulations to analyse the mesoscopic mechanism of chip formation, validated via single-factor cutting experiments. Scanning electron microscopy (SEM) and image processing techniques were used to characterize chip morphology and geometric features, and the mapping relationship between cutting parameters, chip morphology, and surface damage was established. Findings show that serrated chips are easily formed during SiCp/Al cutting; the shape and distribution of SiC particles are key factors regulating chip and surface morphology, with a 6.31% error between simulation and experimental results. With increased cutting speed, chip serration intensifies; enhanced plastic flow of the aluminium matrix drives serrations to evolve into continuous shear flow, with simultaneous improvement in surface quality. As cutting depth increases, chip serration also rises; however, intensified stress concentration in the cutting zone leads to coarser serrations, along with more microcracks, particle pull-outs, and pits, causing a continuous increase in surface roughness. Optimizing cutting parameters can balance chip stability and workpiece surface integrity, providing a process basis for efficient, low-damage machining of SiCp/Al components.

Phenolic resin-impregnated aramid paper honeycombs are promising for engineering applications. However, most designs cannot provide sufficient load-bearing capacity, limiting their application as structural materials. Here, a novel re-entrant honeycomb structure is designed based on the expansion manufacturing method and 17 distinct honeycomb structures are derived from this structure using the Box-Behnken design method. Compression tests are performed on the re-entrant honeycomb to investigate its deformation mechanism and validate the finite element model. Numerical results from 17 honeycomb structures demonstrate that peak stress, plateau stress and energy absorption increase with increasing density and total side length, while decreasing with increasing angle of 1/4 unit cell. The analysis of collapse mechanisms reveal that the plastic deformation mode comprised both bending and membrane deformations. On this basis, an analytical model for predicting the mean compressive strength of honeycomb structures is developed using a simplified super folding element theory. The theoretical solution is in good agreement with the simulations. Furthermore, a multiobjective optimization method is used to efficiently optimize the structural parameters, resulting in the acquisition of optimal structural parameters that meet the load-bearing requirements. The yield strength, mean compressive strength and energy absorption of novel honeycomb are significantly higher than other popular honeycomb structures.

This study prepared three-dimensional orthogonal jute/basalt composites with varying weft mixing ratios via vacuum-assisted resin transfer molding (VARTM). The composites were subsequently immersed in distilled water at two temperatures (25 °C and 65 °C) for up to 90 days.Moisture absorption tests and low-velocity impact (LVI) tests were performed on hygrothermal-treated composites, with damage morphologies characterized using optical microscopy and scanning electron microscopy (SEM). Results showed that moisture absorption rate of composites increased with jute content. Among all specimens, the pure jute composite specimen ([7J]) showed the highest moisture absorption rate, while the pure basalt composite specimen ([7B]) demonstrated the lowest. Elevated temperature accelerated moisture diffusion and matrix degradation, while fiber arrangement also affected moisture absorption performance. Hygrothermal aging degraded LVI performance: after 90-day aging at 65 °C, peak load of the jute/basalt weft-alternating hybrid composite specimen ([4B3J]) specimens decreased by 38.01% and energy absorption reduced by 24%, attributed to matrix hydrolysis, fiber-matrix interface debonding, and microcrack propagation. Damage analysis revealed that specimens with higher jute content experienced more severe fracture and delamination, with aging exacerbating the “I”-shaped back surface damage. This study provides theoretical guidance for engineering applications of natural fiber hybrid composites in hygrothermal environments.

This study investigated impact damage characterization of reinforced thermoplastic pipes (RTP) using X-ray computed tomography (CT) scan and phased array ultrasonic testing (PAUT) techniques. Two types of RTP samples, namely the standard (ST) and gas-tight (GT) types, were impacted using a drop-weight tower. The X-ray CT scan revealed detailed cross-sectional views of damages, including fiber breakage, matrix degradation, and aluminum layer damage. With the implementation of time-corrected gain method in a zero-degree configuration, PAUT demonstrated damage detection capabilities comparable to the CT scan technique by employing various frequencies and focusing techniques. However, it was challenging to accurately assess the extent of damage in the GT type due to the presence of the aluminum layer. While higher-frequency PAUT transducers improved sizing accuracy for the ST type, sizing damages in the GT type remained challenging. Implementing a focusing technique revealed ultrasonic B-scan cross-sectional images of damage closely resembling those from CT scans, offering insights into through-thickness damage morphology. This research showed that the results from the ultrasonic wave-based technique were in good agreement with those from the X-ray imaging-based technique.