The present work proposes a new intelligent framework that integrates ANN with GA and NSGA-II for the predictive modelling and optimization of FDM process. The ANN model is established to develop predictive model between process variables, including layer thickness (LT), nozzle temperature (NT), infill density (ID), and travel speed (TS), and mechanical properties such as tensile, compressive, and flexural strength, achieving a mean absolute error of less than 6
Recently junctionless transistors have gained popularity due to lower fabrication complexity than conventional physically doped transistors. In this work, we have presented gate underlapped GaAs on insulator (GOI) Junctionless(JL) FinFET considering 20nm channel length. The gate underlap engineering has been performed at both the source side and the drain side to improve key short channel effects like leakage current and sub-threshold swing. The DC performances of the proposed device are compared with the conventional Silicon-based SOI JL FinFET and the proposed device provides significant improvement in terms of I_ON (76 I_OFF (3255 times), I_ON / I_OFF ( 13895 times), and SS (5.80 g_m2 , g_m3 ) and voltage intersection points (VIP2, VIP3) of the proposed device are evaluated for a wide range of temperature variations. In addition, the circuit-level performances are also investigated by designing the proposed device-based Current-Starved Voltage Controlled Oscillator (CS VCO) using the Cadence Virtuoso tool. The proposed FinFET-based CS VCO provides a wider tuning range of 99
This paper presents a 16-bit True Random Number Generator (TRNG) based on a Self-Timed Ring (STR) oscillator architecture implemented on an FPGA. The proposed TRNG exploits intrinsic timing jitter and metastability effects of an asynchronous STR to generate high-quality entropy without reliance on an external clock. A three-stage ring oscillator serves as the primary entropy source, while D flip-flops (DFFs) are employed for entropy extraction and stabilization. To further suppress statistical bias and improve output balance, a Von Neumann corrector is incorporated as a lightweight post-processing stage. The design is implemented in Verilog and synthesized using Vivado, demonstrating low power consumption of 0.068 W. Pythonbased statistical tests confirm excellent randomness characteristics, achieving entropy values close to the ideal limit (0.9998). Although The entropy values remain nearly unchanged, the Von Neumann corrector primarily enhances statistical balance and reliability rather than significantly increasing entropy, which is critical for cryptographic robustness. Owing to its clock-independent operation, low power footprint, and resilience to environmental variations, the proposed TRNG is well suited for aerospace systems and AI/ML-enabled embedded platforms.
The spectrum of the Laplacian and the signless Laplacian matrix for a graph product is obtained, where both underlying graphs are regular. As an application of this, we have been able to generate the Kirchhoff Index and Wiener Index and determine the number of spanning trees. Additionally, we derived the conditions necessary for obtaining a Laplacian and a signless Laplacian integral product graph.
Aluminium based nanocomposites are promising candidates for different automotive parts like brake discs, clutch plates, cams and aerospace components like actuator joints, landing gear bushings, rotors, etc., where high strength-to-weight ratio and dry sliding wear resistance are critically important. Since these components frequently get exposed to dry sliding tribological environments like high sliding speed as well as longer sliding distance, understanding their dry sliding wear and friction characteristics is essential for reliable material design and performance of these components. Therefore, the purpose of the current study is to examine the effect of sliding speeds and distances on dry sliding tribological response on LM6-1.5 wt% Si3N4 nanocomposite. Nano-composite is synthesized through ultrasonic assisted stir casting (USC). Microstructural characterizations are evaluated through optical microscopy, FESEM, EDX, XRD and elemental mapping to assess successful incorporation and distribution of Si3N4 Nanoparticles. Pin-on-disc experiment is conducted by using EN31 steel disc as the counterface across sliding speed and distance ranging between 0.25 and 1.25 m/s and 300-3000 m respectively. Incorporation of Si3N4 nanoparticles enhanced wear resistance around 30-40 % compared to base alloy under experimental conditions, indicating improved load-bearing ability and resistance to plastic deformation. Worn surfaces and wear debris are further analyzed through FESEM and EDX to evaluate the primary wear mechanisms. Typical observation of worn surfaces depicts how particle incorporation suppresses delamination as well as adhesion and shifts towards abrasion. This mechanistic change renders a new approach for tailoring material design in tribological applications.
