International Journal of Numerical Modelling-Electronic Networks Devices and Fields最新文献

The Hirota bilinear (HB) is a powerful and widely used technique to find various types of solitons of integrable systems. In this manuscript, we implement HB technique to find bilinear form of a dimensionally reduced (3+1)-dimensional KdV–Calogero–Bogoyavlenskii–Schiff (KdV-CBS) equation at z = x, z = y, and z = t. We present various results for distinct auxiliary function to study kink solitons and its fission and fusion dynamics. The MATLAB-2020 is used to display all the results via 3D and line 2D graphs for appropriate values of parameters. These findings provide a strong new insight into the nonlinear features of the model and lay the foundation for future studies in soliton dynamics and nonlinear events in related systems.

In order to address the challenges of complex process and low precision in traditional device modeling, double hidden layer back propagation neural network (BPNN) are trained using the conjugate gradient (CG) algorithm and the Levenberg–Marquardt (LM) algorithm, the CG-BPNN and LM-BPNN models of small signal for gallium arsenide (GaAs) pseudomorphic high electron mobility transistor (pHEMT) are obtained and analyzed here. At first, the scattering parameters (S-parameters) of GaAs pHEMT are divided into training set and test set randomly. Experimental results show that the CG-BPNN model is better than another S-parameters when predicting ImS₁₂ with mean square error (MSE) of 7.6632e-06, while LM-BPNN model predicts ImS₁₂ with MSE of 2.4672e-06. Meanwhile, the MSE of CG-BPNN model is higher than LM-BPNN model when predicting all the S-parameters. In addition, it shows a smaller fluctuation range for the error curve of LM-BPNN model, which is more stable than the CG-BPNN model. Therefore, the double hidden layer LM-BPNN model is the better choice to characterize the small signal of GaAs pHEMT.

The sine cosine algorithm (SCA), a meta-heuristic optimization method, is used in this study to provide a precise linear and elliptical antenna array design for synthesizing the ideal far-field radiation pattern in the fifth-generation (5G) communication spectrum. The wireless communication system will undergo dramatic changes thanks to the forthcoming 5G technology, which offers exceptionally high data rates, increased capacity, reduced latency, and outstanding service quality. The most important component of 5G communications is an accurate antenna array design for an optimum far-field radiation pattern synthesis with a suppressed sidelobe level (SLL) value and half power beam width (HPBW). While long-distance communication necessitates a low HPBW, the entire side lobe area needs a suppressed SLL to prevent interference. The SCA is used in this case to the optimal feeding currents applied to each array member in the design examples of the concentric circular antenna arrays (CCAA) discussed in this article. It shows the litheness and attainment of the propound algorithm named SCA, chosen CCAAs with three rings and varying amounts of components or antenna array sets those are stated as follows: Set I (4, 6, 8 elements), Set II (8, 10, 12 elements), Set III (6, 12, 18 elements), Set IV (8, 14, 20 elements) with and without the center element are amalgamate. Apply the PSO, Jaya, and SCA optimization algorithms for all four Sets of antenna arrays and compare the attained results; the SLL values achieved by the SCA technique are contrasted with those of other current optimization techniques. The outcomes of all examinations reveal that the SCA algorithm achieved a superior SLL reduction over other optimization techniques.

This article presents a approximation method for linear neutral delay differential equations using Hermite polynomials. The method ensures the unknown function to be found by approximating the first derivative with the help of the finite Hermite series. This method reduces the problem to a linear algebraic system using the first derivative approach, matrix operations and collocation points. Also error of the solution is estimated by constructing an error problem. Numerical examples are solved to explain the method and error estimation technique. The results demonstrate the effectiveness and accuracy of the current study. Calculations have been made using MATLAB.

Binary-code decimal (BCD) to Excess-3 converters (BCD-XS3) can be used as an interface between two systems with different codes, and they can be used to synchronize those systems. Two novel approaches for quantum dot cellular automata BCD-XS3 are suggested, and the design parameters and the amount of energy dissipation are examined. The results have shown that design parameters in two designs of BCD-XS3 have been optimized. The new designs of the BCD-XS3 circuits are simulated, and the proposed design is examined in terms of complexity, latency, and total area. Design parameters have been optimized, showing the reduction in design parameters in the two proposed approaches, which are single layers. The first proposed design uses 119 cells with a delay of 1.5 clock cycles whose number of cells, complexity, and total area are improved by 16.78%, 25%, and 18.18%, respectively, compared with the best previous work. The second proposed design uses 81 cells with a delay of 1 clock. The number of cells, complexity, and total area are improved by 43.35%, 50%, and 50%, respectively, comparing the best previous work. Also, this study investigates the energy consumption in BCD-XS3 for 0.5E_k, 1E_k, and 1.5E_k tunneling energy, which is improved by 31.00%, 33.01%, and 34.68%, respectively, for the first design and 42.59%, 48.74%, and 52.46% for the second design.

This paper focuses on the study and development of an efficient numerical method designed to simulate the radar cross section (RCS) of objects buried in lossy ground and illuminated by a plane wave. The primary objective aligns with the overarching challenge of detecting buried targets in the ground using an airborne radar system. In this scenario, a source antenna illuminates a considered 3D domain, and sensors receive the scattered field from the targets. To enable accurate and efficient simulations, the proposed tool utilizes a Cartesian/unstructured mesh and employs hybrid method that combines two finite volume solvers. In the contents of the paper, we first present a strategy for obtaining Cartesian/unstructured meshes. Subsequently, we study the hybridization of two specific finite volume schemes. Additionaly, a ground and a Near- to Far-field model are introduced for buried targets. To validate and showcase the advantages of our hybrid approach, practical examples are presented. Finally, the strategy designed for handling meshes composed of multiple Cartesian and unstructured zones is detailed.

This paper presents a new approach to particle detection using an active microresonator operating in the transparency regime. Simulations demonstrate that when particles interact with the microresonator surface, they induce optical losses. To compensate for these losses, the optical gain is amplified to restore the transparency regime. Simulation results show a linear relationship between nanoparticle concentration and the pump power required to compensate for optical losses. By the use of microresonator with a very high quality factor, this approach offers an accurate and sensitive method for detecting nanoparticles, without the need for complex equipment.

This paper proposes a new methodology for sensitivity analysis evaluation, fast and with high precision of the electric potential distribution near the transmission lines (TL's). The TL is modeled by the finite element method and the sensitivity of the cable positions is obtained using the adjoint method. The sensitivity of the objective function during the optimization process by using methods based on gradient information is obtained by using the adjoint method. The exact sensitivity obtained by the adjoint method concerning the numeric model of the TL's results in new geometries of the bundles conductors with high surge impedance loading. These geometries are not possible to get using analytic models. The sensitivities of a large number of conductors by phase are obtained with high precision and very low computational cost using adjoint sensitivities, which are independent of the number of design variables. The central finite difference method is used to calculate the sensitivities and to validate the adjoint method. With this methodology, the FEM model of the TL can be used during the optimization process without compromising the required computational processing time.

Physics-informed neural networks (PINNs) provide a new class of mesh-free methods for solving differential equations. However, due to their long training times, PINNs are currently not as competitive as established numerical methods. A promising approach to bridge this gap is transfer learning (TL), that is, reusing the weights and biases of readily trained neural network models to accelerate model training for new learning tasks. This work applies TL to improve the performance of PINNs in the context of magnetostatic field simulation, in particular to resolve boundary value problems with geometrical variations of the computational domain. The suggested TL workflow consists of three steps. (a) A numerical solution based on the finite element method (FEM). (b) A neural network that approximates the FEM solution using standard supervised learning. (c) A PINN initialized with the weights and biases of the pre-trained neural network and further trained using the deep Ritz method. The FEM solution and its neural network-based approximation refer to an computational domain of fixed geometry, while the PINN is trained for a geometrical variation of the domain. The TL workflow is first applied to Poisson's equation on different 2D domains and then to a 2D quadrupole magnet model. Comparisons against randomly initialized PINNs reveal that the performance of TL is ultimately dependent on the type of geometry variation considered, leading to significantly improved convergence rates and training times for some variations, but also to no improvement or even to performance deterioration in other cases.

Complementary metal oxide semiconductor (CMOS) comparators play a pivotal role in analog and mixed-signal circuits, finding diverse applications across electronic systems. In data converter circuits, the significance of high-speed, low-power comparators is pronounced. They ensure swift and precise signal comparisons, minimizing energy usage for dependable analog-to-digital and digital-to-analog conversions. This paper introduces an advanced CMOS dynamic comparator, optimized for data converter circuits using a 45 nm CMOS process. The comparator integrates two novel designs tailored for operation at 0.8 and 1 V power supplies, functioning at 1 GHz. One design incorporates a cascode differential amplifier in the pre-amplifier stage, enhancing speed and sensitivity by augmenting gain, linearity, and output swing. This approach achieves a delay of 73.53 ps and consumes 9.95 μW at a 1 V supply voltage. The second design employs a simple charge pump in the pre-amplifier stage, further elevating speed and sensitivity through amplified voltage levels and enhanced slew rate, resulting in a 57.24 ps delay and 9.03 μW power consumption at 1 V. Simulations underscore the proposed comparator's superiority over conventional counterparts, showcasing significant enhancements in speed and power efficiency, all while preserving precision and dependability.