Delamination and thrust force analysis in GLARE: Influence of tool geometry and prediction with machine learning models DOI
Ergün Ekı̇cı̇, İbrahim Pazarkaya, Gültekin Uzun

et al.

Journal of Composite Materials, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 12, 2024

The multi-layered (fiber/metal) structure of glass fibre aluminium reinforced epoxy (GLARE) makes it difficult to obtain acceptable damage-free holes that meet aerospace standards. This paper investigated the effects tool geometry and drilling parameters on reducing delamination damage uncut fibers at hole exit surface in GLARE. surfaces were examined by scanning electron microscope (SEM) various magnifications. In addition, deep neural network (DNN) long-short-term memory (LSTM) machine learning models used predict (F da ), fiber (UCF), thrust forces using experimental data. No positive contribution special was observed, while standard found be ideal for conditions. Analysis variance (ANOVA) revealed feed rate contributed 57.83% damage, 74.31% 92.33% force, respectively. SEM analysis high deformation zones aluminum layers fracture separation polymer (GFRP) layers. DNN LSTM provide accurate predictions with R 2 values greater than 95% 98%,

Language: Английский

Enhancing machining accuracy of banana fiber-reinforced composites with ensemble machine learning DOI

S. Saravanakumar,

S. Sathiyamurthy,

V. Vinoth

et al.

Measurement, Journal Year: 2024, Volume and Issue: 235, P. 114912 - 114912

Published: May 14, 2024

Language: Английский

Citations

24
Optimized machine learning with hyperparameter tuning and response surface methodology for predicting tribological performance in bio‐composite materials DOI

S. Saravanakumar,

S. Sathiyamurthy,

Vinoth Viswanathan

et al.

Polymer Composites, Journal Year: 2024, Volume and Issue: 45(10), P. 9421 - 9439

Published: April 9, 2024

Abstract This study investigates the influence of NaOH treatment on tribological behavior in hybrid fiber‐reinforced composites, specifically employing Banana fiber with Al 2 O 3 filler an epoxy matrix. Through design experiments (DOE), disc speed, wear duration, and are analyzed for specific rate (SWR) coefficient friction (COF). To advance understanding characteristics, leverages advanced machine learning, using Python‐powered artificial neural networks (ANN), is integrated innovative ANN hyperparameter optimization. Optimized parameters (1050 rpm, 60 s, 5% treatment) significantly minimize SWR (12.38 × 10 −5 mm /Nm) COF (0.2). Scanning electron microscopy (SEM) analysis reveals improved interfacial adhesion identifies micro‐cracks as primary mechanism. work contributes to a profound offering fine‐tuned predictive model optimizing advancing material science engineering. Highlights Reduced SWR, composites via DOE: Explored impact. Advanced learning techniques enhanced prediction. Innovative: Optimal Parameters: 1050 treatment.

Language: Английский

Citations

17
Influence of treatment and fly ash fillers on the mechanical and tribological properties of banana fiber epoxy composites: experimental and ANN-RSM modeling DOI

S. Saravanakumar,

S. Sathiyamurthy,

G. Selvakumar

et al.

Composite Interfaces, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 33

Published: Jan. 6, 2025

This research examines the impact of chemical treatment and banana fly ash fillers on mechanical, tribological, water absorption characteristics fiber-reinforced epoxy composites. Alkaline enhanced fiber-matrix adhesion, markedly improving mechanical characteristics. The optimal performance occurred at 10% content, yielding a tensile strength 40.25 MPa, flexural 77.23 an 44.82 kJ/m2. Water studies indicated decline in moisture uptake, reducing from 40% untreated composites to 25% containing 15% ash, due bonding fewer voids. Tribological experiments demonstrated decrease Specific Wear Rate (SWR) Coefficient Friction (COF) with elevated concentration, signifying improved wear resistance. Predictive modeling Artificial Neural Networks (ANN) showed accuracy (mean error: 0.9584% for SWR, 0.50265% COF). RSM optimization identified input parameters minimizing SWR COF: sliding velocity 5.14491 m/s, distance 652.05 m, content 12.6236%, minimum COF values 15.63 × 10− 5 mm3/Nm 0.242241, respectively. SEM analysis confirmed that treated fibers minimized crack propagation while fracture toughness. results underscore promise ash-filled automotive, aerospace, structural applications necessitate moisture-resistant

Language: Английский

Citations

3
Optimized Machine Learning Prediction and RSM Optimization of Mechanical Properties in Boiled Eggshell Filler-Added Biocomposites DOI
Gopi Periyappillai,

S. Sathiyamurthy,

S. Saravanakumar

et al.

Fibers and Polymers, Journal Year: 2024, Volume and Issue: 25(8), P. 3115 - 3133

Published: July 19, 2024

Language: Английский

Citations

10
Assessment of mechanical, thermal, and sliding wear performance of chemically treated alumina‐filled biocomposites using machine learning and response surface methodology DOI Creative Commons

V.S. Shaisundaram,

S. Saravanakumar,

Rajesh Mohan

et al.

Polymer Composites, Journal Year: 2025, Volume and Issue: unknown

Published: March 14, 2025

Abstract This study examines how NaOH treatment and alumina filler affect the mechanical properties, water absorption, thermal degradation, sliding wear of epoxy composites reinforced with pineapple leaf fiber. greatly improved composites' tensile, flexural, impact strengths by strengthening bond between fiber matrix. Furthermore, incorporation further elevated properties. The composite 10% showed peak values 41.4 MPa in tensile strength, 63.8 flexural 37.6 kJ/m 2 strength. Because hygroscopic parts were removed from treated composites, they absorbed much less water. 15% had lowest absorption at 18% after 192 h. Thermal degradation analysis that stability, having highest char residue (15.3%) 700°C. Sliding tests reinforcement significantly reduced specific rate (SWR) coefficient friction (COF). an SWR 0.2598 × 10 −5 mm 3 /Nm a COF 0.103 when 120 cm/s, 45 N load over 1500 m distance. A scanning electron microscopy found untreated experienced severe abrasive wear, while exhibited mild adhesive wear. shows treating PALF adding enhance their mechanical, thermal, tribological making them suitable for high‐performance industrial applications. Highlights Alumina (41.4 MPa) strength (63.8 MPa). NaOH‐treated moisture, enhancing durability. stability improved, 15.3% 700°C alumina. Optimized achieved (0.2598 /Nm). Artificial neural network response surface methodology accurately predicted optimized behavior.

Language: Английский

Citations

1
ADVANCED ENSEMBLE MACHINE LEARNING PREDICTION TO ENHANCE THE ACCURACY OF ABRASIVE WATERJET MACHINING FOR BIOCOMPOSITES DOI
Gopi Periyappillai,

S. Sathiyamurthy,

S. Saravanakumar

et al.

Materials Chemistry and Physics, Journal Year: 2024, Volume and Issue: unknown, P. 130175 - 130175

Published: Nov. 1, 2024

Language: Английский

Citations

7
Optimization and Finite Element Analysis Simulation on Mechanical Behavior of Wire Arc Additive Manufacturing for SS316L Using Response Surface Methodology DOI

Anantha R. Sethuraman,

Vijayaragavan Elumalai, T. Lakshmanan

et al.

Journal of Materials Engineering and Performance, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 7, 2025

Language: Английский

Citations

0
Machine Learning Prediction and Optimization of Cold Metal Transfer Welding Parameters for Enhancing the Mechanical and Microstructural Properties of Austenitic-Ferritic Stainless-Steel Joints DOI

R. Ravikumar,

A. Mathivanan

Journal of Materials Engineering and Performance, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 14, 2025

Language: Английский

Citations

0
Advanced ensemble machine learning and response surface methodology for optimizing and predicting tribological performance of CMT-WAAM fabricated Al5356 alloy DOI

M.K. Nagarajan,

Manikandan Arumugam

International Journal on Interactive Design and Manufacturing (IJIDeM), Journal Year: 2025, Volume and Issue: unknown

Published: March 20, 2025

Language: Английский

Citations

0
Artificial Intelligence‐Driven Prediction and Optimization of Tensile and Impact Strength in Natural Fiber/Aluminum Oxide Polymer Nanocomposites DOI Creative Commons
Jothi Arunachalam Solairaju, R. Saravanan,

Nashwan Adnan Othman

et al.

Engineering Reports, Journal Year: 2025, Volume and Issue: 7(4)

Published: April 1, 2025

ABSTRACT This study investigates the mechanical properties of hybrid composites reinforced with jute, kenaf, and glass fibers, incorporating Aluminum Oxide (Al 2 O 3 ) as a nanoparticle filler. The effects three key parameters—fiber orientation, fiber sequence, weight percentage Al on—the tensile impact strength were examined. Three levels for each factor considered: orientation (0°, 45°, 90°), sequence (1, 2, layers), varying content (3%, 4%, 5%). response surface methodology (RSM) was employed to optimize parameters, providing insights into interactions between these factors their influence on composite's performance. Additionally, artificial neural networks (ANN) used prediction modeling. outcome presented that ANN model outpaced RSM in terms accuracy, higher correlation predicted experimental values. optimal parameters achieving highest determined, at 90°, 3, 5%. demonstrates effectiveness predicting laminated composite highlights significant role reinforcement enhancing

Language: Английский

Citations

0