Low-density materials, for example, aluminium honeycomb sandwich panels, are gaining extensive usage in structural applications due to their high strength-to-weight ratio and energy absorption capability. This paper presents the investigations on the low-velocity impact behavior of panels reinforced with nano-clay particles at different contents (0%, 2%, and 4% by weight), fabricated by the hand layup method. Panels were tested under impact energies of 10, 20, 30, and 40 J, and surface damage was quantified using MATLAB image processing to obtain numerical damage area (DA). Five machine learning (ML) models were developed to predict DA, with the Polynomial Regression (PR) model (R2 = 0.94) provided a better balance between prediction accuracy and generalization, supported by Cρ (0.90) and LOOCV (R2 ≈ 0.95) analyses. SHAP, Feature importance and Partial dependence analysis confirmed impact energy (IE) as the most influential factor, followed by nano-clay percentage (NCP). Experimentally, 2% NCP was identified as the optimal reinforcement level, achieving maximum damage reduction 41% at 10 J, decreasing to 23% at 40 J while 4% NCP yielded marginal additional benefits. Within the 2% NCP, the damage reduction decreased from 41% at 10 J to 23% at 40 J, which further confirms that impact energy is the most dominant factor, as derived from SHAP, feature importance, and Partial dependence analysis. The ANN model presented moderate success regarding the reconstruction of damage features with PSNR values in the range of 13.30–14.81 dB and SSIM scores between 0.62 and 0.64. Overall, integrating experimental analysis with ML modeling offers a robust route to predicting and understanding the impact-induced damage in composite structures.
HassanHSaeedN. Advancements and applications of lightweight structures: a comprehensive review. Discover Civil Eng2024; 1: 47. https://doi.org/10.1007/s44290-024-00049-z
3.
SunGChenDZhuG, et al.Lightweight hybrid materials and structures for energy absorption: a state-of-the-art review and outlook. Thin-Walled Struct2022; 172: 108760. https://doi.org/10.1016/j.tws.2021.108760
4.
TaubALuoA. Advanced lightweight materials and manufacturing processes for automotive applications. MRS Bull2015; 40: 1045–1054. https://doi.org/10.1557/MRS.2015.268
5.
PhiriRRangappaSSiengchinS, et al.Advances in lightweight composite structures and manufacturing technologies: a comprehensive review. Heliyon2024; 10: e39661. https://doi.org/10.1016/j.heliyon.2024.e39661
6.
SinghJSrivastawaKJanaS, et al.Advancements in lightweight materials for aerospace structures: a comprehensive review. Acceleron Aerospace J2024; 2: 173. https://doi.org/10.61359/11.2106-2409
7.
RajkumarS. Strength and stiffness characteristics of A3003 aluminum honeycomb core sandwich panels. Mater Today Proc2020; 37: 1140–1145. https://doi.org/10.1016/j.matpr.2020.06.348
8.
LiLJiaFLiuL, et al.Response of aluminium honeycomb sandwich panels under combined shock and impact loading: experimental and numerical investigations. Thin-Walled Struct2023; 193: 111256. https://doi.org/10.1016/j.tws.2023.111256
9.
YahayaMRuanDLuG, et al.Response of aluminium honeycomb sandwich panels subjected to foam projectile impact - an experimental study. Int J Impact Eng2015; 75: 100–109. https://doi.org/10.1016/J.IJIMPENG.2014.07.019
10.
JayaramRSNagarajanVAVinod KumarKP. Low velocity impact and compression after impact behaviour of polyester pin-reinforced foam filled honeycomb sandwich panels. J Sandw Struct Mater2022; 24(1): 157–173. https://doi.org/10.1177/1099636221998180
11.
ShiSZhouXZhangJ, et al.In-plane compressive response of composite sandwich panels with local-tight honeycomb cores. Compos Struct2023; 314: 116970. https://doi.org/10.1016/j.compstruct.2023.116970
12.
Abid Allah HamzaNAl-FatlweFTalibM. Experimental and numerical study of effecting core configurations on the static and dynamic behavior of honeycomb plate with aluminum material. Def Technol2023; 39: 177. https://doi.org/10.1016/j.dt.2023.11.016
13.
RaoKKRaoKJSarwadeAG, et al.Bending behavior of aluminum honey comb sandwich panels. Int J Eng Adv Technol2012; 1: 268–272.
14.
ZhangYLiYGuoK, et al.Dynamic mechanical behaviour and energy absorption of aluminium honeycomb sandwich panels under repeated impact loads. Ocean Eng2020; 219: 108344. https://doi.org/10.1016/j.oceaneng.2020.108344
15.
ZhangQZhangXLuG, et al.Ballistic impact behaviors of aluminum alloy sandwich panels with honeycomb cores: an experimental study. J Sandw Struct Mater2018; 20: 861–884. https://doi.org/10.1177/1099636216682166
16.
YanBWangXPanS, et al.Stability and failure of the edge- closed honeycomb sandwich panels with face/core debonding. Appl Sci2020; 10: 7457.
17.
WeiXWuQGaoY, et al.Bending characteristics of allcomposite hexagon honeycomb sandwich beams: experimental tests and a three-dimensional failure mechanism map. Mech Mater2020; 148: 103401.
18.
JamshidiSAzhariMAmoushahiH. Free vibration of plates of various shapes with intermediate point supports by the Hp-Cloud method and lagrange multiplier. Int J Struct Stabil Dynam2016; 16(09): 1550055. https://doi.org/10.1142/S0219455415500558
19.
HosseiniHRAmoushahiHAzhariM. Nonlinear effect of low-velocity impact on tapered laminated composite structures using spline finite strip method. Structures2022; 39: 739–764. https://doi.org/10.1016/j.istruc.2022.03.051
20.
TengXXuYZhangW, et al.The use of pre-hole Z-pinning (PHZ) method to improve the impact resistance and post-impact flexural performance of composite skin/stringer joints. Thin-Walled Struct2022; 180: 109909. https://doi.org/10.1016/j.tws.2022.109909
21.
ZhangXVan HoaSLiY, et al.Effect of Z-pinning on fatigue crack propagation in composite skin/stiffener structures. J Compos Mater2018; 52: 275–285. https://doi.org/10.1177/0021998317710708
MoranoCWagihAAlfanoM, et al.Improving performance of composite/metal T-joints by using corrugated aluminum stiffeners. Compos Struct2022; 307: 116652. https://doi.org/10.1016/j.compstruct.2022.116652
26.
JayaramRSNagarajanVAVinod KumarKP. Compression and low velocity impact response of sandwich panels with polyester pin-reinforced foam filled honeycomb core. J Sandw Struct Mater2019; 21(6): 2014–2030.
27.
FerdousWManaloAAravinthanT. Bond behaviour of composite sandwich panel and epoxy polymer matrix: taguchi design of experiments and theoretical predictions. Constr Build Mater2017; 145: 76–87. https://doi.org/10.1016/J.CONBUILDMAT.2017.03.244
28.
AnbusagarNPalanikumarKGiridharanPK. Study of sandwich effect on nanoclay modified polyester resin GFR face sheet laminates. Compos Struct2015; 125: 336–342.
29.
BozkurtEKayaETano˘gluM. Mechanical and thermal behavior of non-crimp glass fiber reinforced layered clay/epoxy nanocomposites. Compos Sci Technol2007; 67: 3394–3403.
30.
PrasanthPVSahaya RajMEJappesJW, et al.Study of matrix modifications by NC on the mechanical properties of AA3003 honeycomb sandwich panels. Proc IME C J Mech Eng Sci2022; 236(14): 7974–7983.
DinicFSinghKDongT, et al.Applied ML for developing next‐generation functional materials. Adv Funct Mater2021; 31: 2104195. https://doi.org/10.1002/adfm.202104195
35.
Osa-uwagboeNUduAGGhalatiMK, et al.A ML-enabled prediction of damage properties for fiber-reinforced polymer composites under out-of-plane loading. Eng Struct2024; 308: 117970. https://doi.org/10.1016/j.engstruct.2024.117970
36.
DhaneshENagarajanVAVinod KumarKP, et al.Quantitative assessment of impact damage in stitched foam-filled aluminium honeycomb sandwich panels by experimental and ML methods. Polym Compos2024; 45(15): 13663–13675. https://doi.org/10.1002/pc.28726
37.
Ranjbar-RoeintanMAhmadianSSoleymaniA. Predicting low-velocity impact on the CNTs-reinforced plate using ML models. World J Eng2024; 22: 1290.
38.
KamarianSTeimouriAAliniaM, et al.ML for bending behavior of sandwich beams with 3D‐printed core and natural fiber‐reinforced composite face sheets. Polym Compos2024; 45(4): 3043–3054.
39.
QingSLiC. Data-driven prediction on critical mechanical properties of engineered cementitious composites based on ML. Sci Rep2024; 14: 15322. https://doi.org/10.1038/s41598-024-66123-9
40.
ParraDFerreiraGDíazJ, et al.Supervised ML models for mechanical properties prediction in additively manufactured composites. Appl Sci2024; 14(16): 7009. https://doi.org/10.3390/app14167009
41.
VanerioDGuaglianoMBagherifardS. ML image-based analysis for bead geometry prediction in fused granulate fabrication for large format additive manufacturing. NPJ Adv Manuf2025; 2(1): 8.
42.
ManujeshBJPrajnaMR. Damage detection and classification for sandwich composites using ML. Mater Today Proc2022; 52: 702–709.
43.
ViottiIDGomesGF. Delamination identification in sandwich composite structures using ML techniques. Comput Struct2023; 280: 106990.
44.
RautelaMSenthilnathJMonacoE, et al.Delamination prediction in composite panels using unsupervised-feature learning methods with wavelet-enhanced guided wave representations. Compos Struct2022; 291: 115579.
45.
MehrpouyaMGisarioANematollahiM, et al.The prediction model for additively manufacturing of NiTiHf high-temperature shape memory alloy. Mater Today Commun2021; 26: 102022.
46.
WenCZhangYWangC, et al.Machine learning assisted design of high entropy alloys with desired property. Acta Mater2019; 170: 109–117.
47.
ZhangYXuX. Transformation temperature predictions through computational intelligence for NiTi-based shape memory alloys. Shape Mem Superelasticity2020; 6(4): 374–386.
TanchenkoA. Visual-PSNR measure of image quality. J Vis Commun Image Represent2014; 25(5): 874–878.
50.
DeshpandeRGRaghaLLSharmaSK. Video quality assessment through PSNR estimation for different compression standards. Indones J Electr Eng Comput Sci2018; 11(3): 918–924.
51.
LiCBovikAC. Three-component weighted structural similarity index. In Image quality and system performance VI. SPIE, 2009, Vol. 7242, pp. 252–260.
52.
SaraUAkterMUddinMS. Image quality assessment through FSIM, SSIM, MSE and PSNR—a comparative study. J Comput Commun2019; 7(3): 8–18.
53.
SchoberPBoerCSchwarteL. Correlation coefficients: appropriate use and interpretation. Anesth Analg2018; 126: 1763–1768.
54.
DiagnosticOnL. Pearson's correlation coefficient. N Z Med J1996; 109(1015): 38.
55.
BerggrenM. Coefficients of determination measured on the same scale as the outcome: alternatives to R2 that use standard deviations instead of explained variace. Psychol Methods2024, Advance online publication.
56.
NordinNZainolZNoorM, et al.An explainable predictive model for suicide attempt risk using an ensemble learning and Shapley Additive Explanations (SHAP) approach. Asian J Psychiatry2022; 79: 103316. https://doi.org/10.1016/j.ajp.2022.103316
57.
OralBMaddahHYıldırımR. Predictive modeling and SHAP (SHapley Additive ExPlanations) analysis for enhancing natural dye‐sensitized solar cell performance. Sol RRL2024. https://doi.org/10.1002/solr.202400432
