ABSTRACT Improving the fatigue life of advanced composites and accurately predicting their fatigue life are crucial because composites are increasingly used in applications involving repeated loading, including automotive chassis, wind turbine blades, aircraft structures, and bridges. Graphene‐modified glass fiber‐reinforced polymer (GFRP) composites were fabricated using the vacuum resin infusion method. The fatigue life of graphene‐modified GFRP laminates was improved by about three to five times by modifying the epoxy resin with 0.5% graphene, compared to GFRP without graphene. Moreover, the fatigue strength coefficient and ultimate tensile strength of graphene‐modified GFRP were enhanced by about 38% and 27% compared to neat GFRP. Fatigue life prediction in composites, particularly nanomaterial‐reinforced systems, is challenging due to pronounced anisotropy and progressive damage evolution governed by reinforcement type and configuration. Moreover, the limited availability of fatigue data for nanomaterial‐reinforced composites presents challenges for conventional data‐driven prediction. Therefore, a novel physics‐guided neural network (PGNN) framework was proposed under small sample conditions for accurate fatigue‐life prediction of graphene‐modified GFRP composites. The PGNN exhibited superior predictive performance relative to the purely data‐driven model, achieving an R 2 of 0.9804 compared with 0.8900, while the RMSE decreased from 0.3449 to 0.1453 and the MAE from 0.0860 to 0.0747 on the log‐transformed scale.
ABSTRACT This study investigates the low‐velocity impact behavior of 5052‐H32 aluminum‐based fiber metal laminates (FMLs) under varying impact energies, fiber types, and short‐term NaCl exposure. A full‐factorial experimental design was employed using three FML configurations (GLARE, BARAL, and HGBAL), three impact energies, and two exposure conditions. Drop‐weight impact tests were conducted, and the responses were evaluated using force–time, force–displacement, and energy–time curves, together with analysis of variance (ANOVA). Increasing impact energy increased peak force, permanent displacement, and final absorbed energy for all laminates. The increase in permanent deformation indicates greater irreversible deformation at higher impact energies; however, the three energy levels do not allow a specific transition threshold to be established. GLARE generally exhibited a relatively stiffer initial response and lower permanent deformation, whereas BARAL and HGBAL showed greater permanent deformation at higher impact energies. Short‐term NaCl exposure changed peak force and deformation behavior, while final absorbed energy and energy absorption efficiency remained close to those of unexposed laminates. ANOVA showed that impact energy was the dominant factor affecting peak force ( p < 0.001), while the fiber type and NaCl exposure interaction was also statistically significant ( p = 0.004).
ABSTRACT This study develops a micromechanical and numerical framework to predict the effective elastic properties of coconut shell powder (CSP)‐reinforced epoxy composites. Classical analytical homogenization models are used to systematically examine the effects of reinforcement volume fraction, particle geometry, orientation distribution, and porosity, while finite element simulations on periodic representative volume elements (RVEs) are used for the spherical‐particle reference configuration. Particle geometry and orientation are shown to govern the symmetry of the effective stiffness tensor. The predicted Young's modulus of the four‐phase configuration increases from 4.23 GPa for neat epoxy to approximately 5.15 GPa at 50 wt% CSP, corresponding to a volume fraction of 48.61%. Porosity is represented by a zero‐stiffness void phase. A progressive multi‐phase Mori–Tanaka framework is then constructed by successively incorporating an orientation distribution function (ODF)‐based orientation description, bimodal morphological heterogeneity through coexisting spherical and ellipsoidal inclusion families, and an explicit porosity phase. Among the three Mori–Tanaka variants investigated, the four‐phase model shows the closest overall agreement with the experimental reference data, with an estimated mean absolute relative error of 3.55%. Because the CSP constituent properties were derived from the composite data used for comparison, this agreement reflects internal consistency rather than independent validation.
ABSTRACT A polymethyl methacrylate/methyl methacrylate (PMMA/MMA) binary liquid resin (PMBLR) was employed to fabricate polyimide‐, aramid‐, and ultrahigh‐molecular‐weight polyethylene (UHMWPE)‐fabric‐reinforced PMMA composites via vacuum‐assisted resin infusion (VARI). The low porosity of polyimide/PMMA (0.171%), aramid/PMMA (0.251%), and UHMWPE/PMMA (0.058%) demonstrated effective impregnation of the three fabrics by the PMMA/MMA resin. Mechanical testing showed that the tensile strengths of polyimide/PMMA, aramid/PMMA, and UHMWPE/PMMA decreased from 728 to 623 MPa, from 667 to 554 MPa, and from 485 to 409 MPa, respectively, as the temperature increased from −30°C to +50°C, corresponding to reductions of 14%, 17%, and 16%. The three composites showed comparable short‐beam strengths (14.37–14.76 MPa), but Polyimide/PMMA exhibited a stiffness‐dominated, high‐load‐bearing response, with the highest flexural strength (206 MPa) and impact strength (233 kJ/m 2 ). Its flexural and impact strengths were 17% and 28% higher than those of Aramid/PMMA, and 137% and 36% higher than those of UHMWPE/PMMA, respectively. Aramid/PMMA exhibited intermediate mechanical performance, whereas UHMWPE/PMMA showed a more compliant response with lower flexural and impact strengths. MMA‐assisted dissolution removed the PMMA matrix while preserving the macroscopic integrity of the organic fabrics, supporting the feasibility of fabric recovery from these PMMA‐based composites.
ABSTRACT Ultrasonic welding of eight‐harness satin woven glass‐fiber‐reinforced polyphenylene sulfide (GF/PPS) composites was investigated by comparing time‐, energy‐, and displacement‐controlled modes and examining the coupled effects of vibration amplitude, welding force, and normalized sonotrode travel. Lap‐shear testing, power–time analysis, metallography, micro‐computed tomography (micro‐CT), and scanning electron microscopy (SEM) were combined to investigate process–structure–property relationships. Nine representative SEM fracture fields were quantitatively classified into Pore/void, Residual resin, and Broken fibers regions. Their projected area fractions were related to mean lap‐shear strength (LSS) alongside micro‐CT‐derived internal porosity. Joint strength depended on the combined welding parameters, reaching 21.17 ± 0.66 MPa for the strongest investigated 90° condition. At matched normalized travel, lower representative internal porosity was consistently associated with higher LSS. Greater areas of fiber‐fracture features and smaller pore/void regions on the fracture surfaces also corresponded to improved joint performance, reflecting changes in the involvement of adjacent fiber‐reinforced regions during failure. These findings clarify interfacial consolidation, defect evolution, and fracture behavior and provide a basis for subsequent process‐window optimization and quantitative weld‐quality assessment of woven GF/PPS composites.
