
This study looks at the characterization and performance of ultrasonic welding for ABS-PC (Mychril) polymer blends used in EV station applications. Experiments were conducted and a detailed characterization of the welded interfaces were performed using Scanning Electron Microscopy (SEM) to examine their structure. Fourier Transform Infrared Spectroscopy (FTIR) was deployed to identify chemical interactions and any potential degradation, and X-ray Diffraction (XRD) to evaluate changes in crystallinity caused by the welding process. SEM images showed consistent fusion and minimal voids, while FTIR and XRD tests showed that key functional groups remained intact with slight changes in crystallinity. Mechanical tests were also carried out on the welded samples that involved tensile, impact and fracture assessments. Pearson’s heat map coefficient analysis was performed to understand the influence of input process parameters on the mechanical strength outcomes. This research shows that ultrasonic welding is a suitable, sustainable, and effective method for assembling ABS-PC samples in EV applications, providing mechanical reliability and design flexibility.
The accurate prediction of mechanical properties and defect formation in friction stir welding (FSW) of aluminium alloys is crucial for ensuring structural integrity and reliability in aerospace and automotive applications. In this study, an artificial neural network (ANN) was developed to predict yield strength (YS) and welding defects (WD) in FSW joints of AA2024, AA2219, and AA6061 alloys. A dataset of one hundred experimental cases obtained from peer-reviewed literature was employed, where process parameters rotational speed (RS), plate thickness (PT), shoulder radius (SR), axial pressure (AP), pin root radius (PR), pin tip radius (PS), and tool angle (TA) were served as inputs, and YS and WD were target outputs. The ANN was implemented in MATLAB R2016a and trained using backpropagation. Results showed strong predictive accuracy: for YS, correlation coefficients (R) of 0.96568, 0.99874, and 0.96278 were achieved for training, validation, and testing sets, with an overall R of 0.96764. The minimum validation error (MSE = 2.1901) occurred at epoch 11. For WD prediction, the overall R was 0.83229, with lowest validation error (MSE = 0.09041) at epoch 17. These findings highlight the potential of ANN-based models for real-time prediction and optimization of FSW quality. Future work will focus on expanding datasets, integrating hybrid AI techniques, and developing adaptive models for industrial-scale welding applications.
Electrical Discharge Machining (EDM) is a non-traditional machining process which employed to make complex products from hard materials. The machining efficiency and surface integrity are influenced by the selection of parameters. In this research, datasets were collected from earlier experimental research works on EDM process of AL alloy to evaluate machining performance. The input parameters: pulse-on time, pulse-off time, discharge current, gap voltage, flushing pressure, and tool rotational speed and the output responses: material removal rate (MRR), tool wear rate (TWR), surface roughness (SRS), recast layer (RLR), and microhardness (MHS) were considered. MINITAB software was utilized to identify the most significant factors affecting responses through signal-to-noise analysis. For multi-objective optimization, the Taguchi method was integrated with Grey Relational Analysis (GRA) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). From the results, GRA identified optimal parameters of discharge current of 11 A, pulse-on time of 75 µs, pulse-off time of 25 µs, gap voltage of 50 V, flushing pressure of 0.4 MPa, and tool speed of 600 rpm, yielding the highest productivity with MRR of 4.76 g/min, moderate TWR of 0.478 g/min. In contrast, TOPSIS suggested a discharge current of 7 A, pulse-on time of 175 µs, pulse-off time of 90 µs, gap voltage of 40 V, a flushing pressure of 0.5 MPa, and a tool speed of 900 rpm, which produced superior surface quality with a lower SRS of 7.34 µm, reduced RLR of 19.6 µm, and higher MHS of 120.9 HV and a reduced MRR of 0.78 g/min. Both optimal results were validated using an Adaptive Neuro-Fuzzy Inference System (ANFIS) model, confirming accurate prediction of EDM responses. This study demonstrates that GRA is more suitable for productivity-focused applications, whereas TOPSIS is advantageous when surface integrity and hardness are critical, offering a robust decision-making framework for EDM optimization.
This work provides a detailed comparative study on the mechanical performance and microstructural features of dissimilar joints produced by Tungsten Inert Gas (TIG) welding between Inconel 718 and AISI 410 martensitic stainless steel. Welding was performed on 2 mm thick sheets of both alloys using optimized conditions: 80 A current, 12.1 V voltage, 60 cm/min travel speed. Microstructural evaluation revealed contrasting solidification patterns, with Inconel 718 exhibiting Laves phase precipitation within dendritic networks, while AISI 410 displayed coarse dendrites interspersed with sulfur-based inclusions. Vickers microhardness analysis indicated a non-uniform distribution across the weldment. The highest hardness (410 HV) occurred in the AISI 410 heat-affected zone (HAZ) due to martensitic transformation, whereas the Inconel 718 base metal and fusion zone recorded 267±3 HV and 264±5 HV, respectively. Tensile testing demonstrated an ultimate tensile strength of 461±2 MPa, a yield strength of 273 MPa, and an elongation of 17.95±0.8% for the welded joint.
In recent decades, due to their great advantages, the aluminium alloys have been more and more used in the shipbuilding industry, particularly in the construction of yacht superstructures. Nevertheless, a significant disadvantage of aluminium alloys subjected to welding is their high deformation tendency, which represents an important drawback that needs additional remedial measures. In this study, a finite element analysis was performed, in order to investigate and optimize the design of welding paths used to carry out AA 5083 aluminium alloy T-joints of 6mm thick sheets by Metal Inert Gas (MIG) welding. The research methodology was focused on three case studies, in order to determine and comparatively analyse the Heat Affected Zone extend and the level of stress and displacement. In the first case, the welding beads were successively deposited, following the same welding direction. In the second case, the seams were successively welded, but the welding directions on one side and on the other side of the joint were opposite. In the third case, the seams were simultaneously welded, on both sides of the joint, the welding directions being opposite, as in the previous case study. Based on the research results, it was found that the larger Heat Affected Zone (HAZ) was identified in the third case study, while the total displacement was 25-30% lower, in comparison with the results obtained in the first two case studies. This phenomenon can be explained by the high amount of heat developed simultaneously by two electric arcs and transferred, by conduction, to the base materials. Due to the joining technology applied, that consisted in welding simultaneously on the both sides of the joint, an adequate balance of stress and strain was achieved, determining a lower total displacement.