Jayavelu Udaya Prakash

Background: Aluminium based composites with hybrid reinforcement hold significant potential to replace Al-alloys in a variety of automotive sectors where cheap cost, a significant ratio of strength to weight, and better wear resistance are required. Methods: Stir casting was utilized to make aluminium matrix composites (AMCs) with 3%, 6%, and 9% of B₄C/Fly ash particles. The wear was examined with various Sliding Speed, S (1 m/s, 1.5 m/s and 2 m/s), Sliding Distance, D (500 m, 1000 m and 1500 m), applied load, L (15 N, 30N and 45 N) and reinforcement %, R (3, 6 and 9%). Grey Relational Analysis was used to optimise the wear variables. Taguchi’s L₂₇ Orthogonal array (OA) was selected for this statistical approach in order to analyse responses like Specific wear rate (SWR) and Coefficient of Friction (CoF). Furthermore, analysis of variance (ANOVA) was utilized to investigate the influence of input parameters on wear behavior by choosing “smaller is better” feature. Results: Based on this study, the optimal values of S – 1.5 m/s, D – 500 m, L – 30 N, and R% – 9 wt% Hybrid (4.5% Fly ash and 4.5% B₄C) are found to yield the lowest SWR and CoF. Wear rate of composite decreased with an increase in reinforcement particles. Increase in hardness was also the reason for decrease in wear rate. The responses have a narrow margin of error, according to confirmation studies. There exists a good agreement between them. Discussion: The research on LM6/B₄C/fly ash composite fabrication using Grey Relational Analysis (GRA) has significantly contributed to the development of high-performance materials for wear-related applications. Through the optimization of wear parameters, GRA allows for the improvement of wear resistance, strength, and sustainability.

LM5 alloy is suitable for metal castings for marine and aesthetic uses due to its admirable resistance to corrosion. In order to make intricate shapes in the LM5 alloy, this study intends to assess the impact of Wire Electric Discharge Machining process variables, like Pulse on Time (T_on), Pulse off Time (T_off), Gap Voltage (GV) and Wire Feed (WF) on responses like Material Removal Rate (MRR), Surface Roughness (SR), and Kerf Width (K_w). The LM5 aluminium alloy plate was produced through stir casting process. SEM, EDAX and XRD images confirm the LM5 Al alloy’s microstructure and crystal structure. WEDM studies were conducted using design of experiments approach based on L₉ orthogonal array and analysed using Taguchi’s Signal to Noise Ratio (S/N) analysis. Pulse on Time has the greatest statistical effects on MRR (68.25%), SR (79.46%) and kerf (81.97%). In order to assess the surface integrity of the WEDM machined surfaces, the SEM study on the topography was conducted using the optimum surface roughness process variables: T_on 110 μs, T_off 50 μs, GV 40 V, and WF 9 m/min. SEM images show the recast layer and its thickness. The average absolute error for MRR is 1.69%, SR is 3.89% and kerf is 0.88%, based on mathematical (linear regression) models. The Taguchi’s Signal to Noise ratio analysis is the most appropriate for single objective optimization of responses.

This paper proposes a convolutional neural network (CNN) model of the signal distribution control algorithm (SDCA) to maximize the dynamic vehicular traffic signal flow for each junction phase. The aim of the proposed algorithm is to determine the reward value and new state. It deconstructs the routing components of the current multi-directional queuing system (MDQS) architecture to identify optimal policies for every traffic scenario. Initially, the state value is divided into a function value and a parameter value. Combining these two scenarios updates the resulting optimized state value. Ultimately, an analogous criterion is developed for the current dataset. Next, the error or loss value for the present scenario is computed. Furthermore, utilizing the Deep Q-learning methodology with a quad agent enhances previous study discoveries. The recommended method outperforms all other traditional approaches in effectively optimizing traffic signal timing.

With the enhancement in science and technology, necessity of complex shapes in manufacturing industries have become essential for more versatile applications. This leads to the demand for lightweight and durable materials for applications in aerospace, defense, automotive, as well as sports and thermal management. Wire electric discharge machining (WEDM) is an extensively utilized process that is used for the exact and indented shaped components of all materials that are electrically conductive. This technique is suitable in practically all industrial sectors owing to its widespread application. The present investigation explores WEDM for LM6/fly ash composites to optimize different process variables for attaining performance measures in terms of maximum material removal rate (MRR) and minimum surface roughness (SR). Taguchi’s L ₂₇ OA design of experiments, grey relational analysis, and analysis of variance (ANOVA) were employed to optimize SR and MRR. It has been noted from ANOVA that reinforcement ( R ) percentage and pulse on time are the most influential aspects for Grey Relational Grade (GRG) with their contributions of 28.22 and 18.18%, respectively. It is found that the best process variables for achieving the highest MRR and lowest SR simultaneously during the machining of the composite are gap voltage of 30 V, pulse on time of 10 µs, pulse off time of 2 µs, wire feed of 8 m/min, and R of 9%. The predicted GRG is 0.84, and the experimental GRG value is 0.86. The validation experiments at the optimized setting show close agreement between predicted and experimental values. The morphological study by optical microscopy revealed a homogenous distribution of reinforcement in the matrix which enhances the composite’s hardness and decreases the density.

The materials used in engineering have seen a significant transformation in the contemporary world. Numerous composites are employed to overcome these problems because conventional materials are unable to meet the needs of current applications. For quite some time, professional engineers and researchers have been captivated by the problem of choosing the best machining parameters for new composite materials. Wire electrical discharge machining is a popular unconventional machining process that is often used for making complex shapes. Numerous process parameters influence the WEDM process. Thus, to achieve affordable and high-quality machining, the right set of process parameters must be provided. Finding the wire cut EDM optimized settings for the fabricated LM5/ZrO₂/Gr composite is the main aim of this research. The chosen input parameters are the wire feed, pulse on and pulse off times, the gap voltage, and the reinforcing percentage. In this study, LM5/ZrO₂/Gr composites were made from stir casting with 6-weight percent ZrO₂ as the reinforcement and varying graphite percentages of 2, 3, and 4 wt%. Then they were machined in WEDM using L₂₇ OA to seek the best parameters for machining by adjusting the input parameters. The findings were analysed by means of grey relation analysis (GRA) to achieve the supreme material removal rate (MRR), lowest surface roughness (SR), and a smaller kerf width (K_w) simultaneously. GRA determines the impact of the machining variables on the standard characteristics and tests the impact of the machining parameters. Confirmation experiments were performed finally to acquire the best findings. The experimental findings and GRA show that the ideal process conditions for achieving the highest grey relational grade (GRG) are 6% ZrO₂ with 2% graphite reinforcement, a wire feed of 6 m/min, a pulse off time (T_off) of 40 µs, a pulse on time (T_on) of 110 µs, and a gap voltage (GV) of 20 V. The gap voltage (22.87%) has the greatest impact on the GRG according to analysis of variance (ANOVA), subsequent to the interaction between the pulse on time and the gap voltage (16.73%), pulse on time (15.28%), and pulse off time (14.42%). The predicted value of the GRG is 0.679; however, the experimental GRG value is 0.672. The values are well-aligned between the expected and the experimental results. The error is only 3.29%, which is really little. Finally, mathematical models were created for each response.