Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Vagheesan Senthilkumar has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Sustainable and Intelligent Automation Framework for Aerospace Alloys Using Multi-Agent Deep Reinforcement Learning

    by Nagadeepan Anbazhagan, Senthilkumar Vagheesan, K. K. Ilavenil

    Advanced aerospace alloys such as titanium alloy (Ti-6Al-4V) are widely employed in critical applications owing to their excellent strength-to-weight ratio and corrosion resistance. However, machining these alloys remains challenging due to significant tool wear, poor material removal rates, and surface integrity concerns. This research offers a sustainable and intelligent machining framework for a 3 mm thick Ti-6Al-4V alloy, utilizing coated wire electrical discharge machining (WEDM) linked with Multi-Agent Deep Reinforcement Learning (MADRL). A Box–Behnken experimental design was adopted to gather baseline data for pulse-on time, pulse-off time, servo voltage, and peak current. The MADRL architecture combines cooperative agents to improve MRR, surface roughness, and kerf width concurrently. Beyond performance increase, the sustainability parameters of energy consumption, dielectric fluid use, and wire consumption were also studied. The proposed MADRL significantly improved the material removal rate (MRR) from 1.00 to 1.28 mm3/min and reduced the average surface roughness (Ra) from 1.95 to 1.66 µm, the kerf width from 0.27 to 0.24 mm, and the energy consumption from 122 to 107 J. The results show the potential of the proposed framework for adaptive, sustainable, and high-performance aerospace manufacturing.

  • Open Access

    Micro Plastic Contamination: A Comprehensive Review of Risks and Sustainable Solutions

    by V. Senthilkumar, M. Anu, P. Sameerabanu, K. K. Ilavenil

    Plastic particles smaller than 5 mm, known as microplastics, come from microbeads, synthetic fibres, and the decomposition of bigger plastics. Surface runoff, wastewater effluent, and air deposition carry these particles to aquatic ecosystems. Microplastics can damage aquatic species, impair feeding and reproduction, and bioaccumulate toxins. Humans may have digestive, respiratory, and reproductive difficulties from intake or inhalation. Bans on microbeads, waste management improvements, sophisticated wastewater treatment plant filtration systems, and consumer education on plastic reduction and disposal are mitigation options. The detection, quantification, and regulation of microplastics remain difficult despite these efforts.


  • Adaptive Optimization of Wire EDM Process for Ti6Al4V Alloy Using ANFIS and AI Technique

    by V. Senthilkumar, A. Nagadeepan

    Wire Electrical Discharge Machining (WEDM) is a non-traditional machining process widely used for machining difficult-to-cut materials such as titanium alloys. This study proposes an intelligent optimization approach using Adaptive Neuro-Fuzzy Inference System (ANFIS) to optimize WEDM parameters for machining titanium alloy (Ti6Al4V) with coated wires. The key parameters investigated include pulse-on time, pulse-off time, peak current, and wire tension. The results demonstrate that the ANFIS model accurately predicts the optimal parameters, achieving a material removal rate (MRR) of up to 5.22 mm3/min (increased by 15%) and a surface roughness (Ra) as low as 3.60 µm (reduced by 12%). The proposed approach significantly improves machining efficiency and surface quality, reducing the need for costly experimental trials. This study highlights the potential of ANFIS in optimizing WEDM processes for industrial applications, particularly in aerospace and biomedical industries where titanium alloys are extensively used.


  • MULTI-CRITERIA OPTIMIZATION OF CO2 LASER CUTTING OF METALS USING HYBRID AHP–TOPSIS APPROACH

    by A. Nagadeepan, M. Pranav, P. Santhosh, S. SivaShankaran, V. Senthilkumar

    The non-linear effects of the laser cutting process parameters and their interactions on the cut quality for ferrous and non-ferrous metals were difficult to predict. It is vital and complex to find the optimum process condition for a specific application and requires evaluation of a number of competing and distinct process performance characteristics. Various Multi Criterion Decision-Making (MCDM) techniques which are simple and logical are available to aid the selection of optimal combination of cutting parameters in the modern manufacturing processes. The laser cutting is one modern manufacturing process which is capable of cutting complex shapes in almost all the engineering materials and requires a variety of parameters and performance characteristics. To predict the non-linear effects in laser cutting, box-behnken design with three process parameters laser beam power, cutting speed and gas pressure was employed to design the experiments. The analytic hierarchy process (AHP) is used to predict the weightage among the responses and then technique for order of preference by similarity to ideal solution (TOPSIS) is used to optimize the process parameters associated with laser cutting of ferrous and non-ferrous alloys

  • AI-BASED PREDICTION OF KERF WIDTH AND SURFACE ROUGHNESS IN CO2 LASER CUTTING OF STAINLESS STEEL

    by A. Nagadeepan, C.S. Tamil Selvan, S. Viswanathan, B. Vishnu, V. Senthilkumar

    Light Amplification by Stimulated Emission of Radiation or LASER, is a thermal energy-based unconventional machining method. CO2 laser cutting of AISI 314 Stainless steel is carried out to evaluate the variation of kerf width in the cut section. Back-propagation Artificial Neural Networks are used to analyse and predict the kerf width during CO2 laser cutting. In this study, input parameters considered were cutting speed, power, stand off distance and gas pressure. For experimental database of artificial neural network L16 taguchi orthogonal array with four levels for each input parameter was proposed. Among the 16 datasets, 12 datasets were used for training the network and the remaining 4 datasets were used for testing the network. The results of predicted roughness and kerf width by back propagation neural network were compared with experimental data and the average predicting error on training datas was 0.37% and the average predicting error on testing datas was 4.34%, which confirms that the predicted ANN model might be utilised to study the impact of CO2 laser cutting settings on kerf width.

  • Open Access

    Investigating and Multi-Objective Optimizing WEDM Parameters for Al6061/Mg/MoS2 Composites Using BBD and NSGA-II

    by Vagheesan Senthilkumar, Anbazhagan Nagadeepan, K. K. Ilavenil

    This study aims to optimize the Wire Electrical Discharge Machining (EDM) process parameters for aluminum 6061 alloy reinforced with Mg and MoS2 using the Box–Behnken (BBD) design and the non-dominated sorting genetic (NSGA-II) algorithm. The objective is to enhance the machining efficiency and quality of the composite material. The Box–Behnken (BBD) design was utilized to design a set of experiments with varying levels of process parameters, comprising pulse-on time, servo volt, and current. The material removal rate and surface roughness were considered as machining responses for optimization. These responses were measured and used to develop a mathematical model. The NSGA-II, a multi-objective optimization algorithm, was then applied to search for the optimal combination of process parameters that simultaneously maximizes the material removal rate and minimizes the electrode wear rate and surface roughness. The algorithm generated and evolved a set of Pareto-optimal solutions, providing a trade-off between conflicting objectives. The results of the optimization process were analyzed to identify the optimal process parameters that lead to improved machining performance. The study revealed optimal Wire Electrical Discharge Machining (WEDM) parameters for Al6061/Mg/MoS2 composites using NSGA-II. The optimized parameters, including a pulse-on time (Ton) of 105 µs, servo voltage (SV) of 35 V, and peak current (PC) of 31 A, resulted in a Material Removal Rate (MRR) of 7.51 mm3/min and a surface roughness (SR) of 1.97 µm. This represents a 15% improvement in the MRR and a 20% reduction in the SR compared to non-optimized settings, demonstrating the efficiency of the BBD-NSGA-II approach.