Transition metal borides with a tetragonal structure are emerging as a strong candidate for rare-earth-free permanent magnet applications due to their inherently high magnetization, magnetocrystalline anisotropy, and Curie temperature. In this study, we developed (Fe0.7-xCo0.3Zrx)2 B (x = 0, 0.01, 0.02, 0.03, and 0.04) hard magnetic phase through arc melting followed by rapid solidification using melt spinning. X-ray diffraction (XRD) confirms the formation of a single phase tetragonal structure, with an average crystallite size in the range of ∼30-40 nm. Singular point detection (SPD) measurement showed that Zr incorporation enhances magnetocrystalline anisotropy, increasing the anisotropy field (Ha) from 7.5 to 9.5 kOe with an increase of Zr content from x = 0 to x = 0.03. Room-temperature magnetization loops reveal saturation magnetization (Ms) of 139–144 emu/g and an optimum coercivity of approximately 0.39 kOe. Additionally, Bloch’s law fitting of temperature-dependent saturation magnetization data suggests weakened ferromagnetic exchange interactions with increasing Zr content. These findings demonstrate that minor Zr incorporation significantly improves magnetic anisotropy, suggesting (Fe0.7Co0.3)2B-based alloys as promising candidates for next-generation, high-performance permanent magnets.

This study proposes a deep learning-based method for the performance evaluation and optimization of recycled concrete pile foundations. By integrating the Particle Swarm Optimization (PSO) algorithm and the Simulated Annealing (SA) algorithm to improve the performance of the BP neural network, a hybrid SA–PSO–BP prediction model is constructed to achieve the prediction of vertical bearing capacity and settlement of recycled concrete pile foundations. Case studies demonstrate that the SA–PSO–BP model controls the prediction errors for pile foundation bearing capacity and settlement within ±8%, outperforming other models. This model excels particularly in capturing time-dependent characteristics under long-term loads, providing a reliable intelligent decision-making tool for the design optimization and safety assessment of pile foundation engineering.

Reflection-based lensing is used to focus or collimate extreme ultraviolet (EUV)/x rays (x-ray optics) propagating in grazing incidence small-angle x-ray scattering (GISAXS) inside poly-capillary glass. In the process, the refractive index of silica is slightly less than that of vacuum (1.0) for EUV/x rays, leading to total external reflection with a critical angle slightly less than 90°. Here, we examine the feasibility of widening the total external reflection window by coating the capillary interior surface with a silicon nanofilm that assumes the shape of the polycapillary, which may enhance the collection and collimation power. We use slow dynamic self-assembly via circulation of the nanoparticles through the capillary. Electromagnetic scattering computations are conducted using the model of a flat glass substrate decorated with closely packed Si nanoparticles. Optical, luminescent, and SEM imaging techniques, as well as spectroscopy, show that the polycapillary coating is uniform and luminescent under UV irradiation. The computations show that, for visible wavelengths, a 3 nm silicon nanoparticle film reduces the grazing reflectivity at the visible/UV band-to-band transitions E1 and E2. For EUV/x ray, there is a significant reflectivity enhancement of ∼10% due to a larger index dip below that of vacuum, affording a smaller critical angle. Widening the external reflection regime promises superior collection and focusing power, improving the intensity and resolution. Being luminescent under UV makes it useful for imaging, therapy, and other clinical procedures, while also enabling the development of more compact and versatile instruments.

A reliable bond between steel reinforcement and recycled aggregate concrete is essential for ensuring composite action and structural integrity. However, flexural pull-out test data that realistically capture this interfacial behavior remain limited, and existing bond strength models often neglect several key parameters. Research on anchorage length design is also relatively scarce. To address these limitations, this study adopts a data-driven framework to compile and analyze available experimental evidence, aiming to develop a reliable predictive model for peak bond strength and to propose an anchorage length formula with appropriate safety margins. A total of 158 test records from ten representative studies were collected under comparable specimen preparation and loading conditions to establish a baseline database. To mitigate data sparsity, advanced data augmentation techniques were employed to expand the dataset to 500 samples while preserving physical boundary constraints, statistical characteristics, and intrinsic correlations, thereby compensating for missing variables and enhancing data diversity. Based on the enriched dataset, five ensemble learning algorithms combined with interpretability tools were used to predict and interpret bond strength, and a reliability analysis was conducted to evaluate anchorage performance. The results indicate that bond strength is primarily governed by the ratio of bar diameter to bond length, cover thickness, recycled coarse aggregate content and quality, matrix strength, stirrup parameters, and recycled fine aggregate properties. All variables exhibit positive correlations except those related to recycled aggregates. Among the models, gradient boosting achieved the highest predictive accuracy without overfitting, while reliability analysis confirmed representative models provide adequate safety margins.

A high-performance MRI based on high-Tc SQUID and flux coupling at ultralow magnetic fields is reported in this work. The technique of NMR/MRI was optimized to study MRI of water phantoms and fingers. An SNR of 62 in image contrast was achieved for a water column of 200 μL at magnetic fields of 98 μT. The MRI showed high image contrast and demonstrated a spatial resolution of 2 mm for samples in magnetically unshielded environments. This resolution is comparable to the current state-of-the-art values (typically 2.0–3.0 mm) reported for SQUID-based ULF MRI systems. A clear image of fingers was demonstrated for the first time using high-Tc SQUIDs, which shows promise for biological or clinical applications.

This paper presents an analytical method based on the Fourier transform and mode-matching technique to evaluate electromagnetic field leakage from long power cables. The proposed method models the two-dimensional transverse cross section of high-power cables and accurately predicts the effects of slot size, slot position, enclosure dimensions, and multiple cable configurations on low-frequency magnetic field penetration. The analytical results show excellent agreement with three-dimensional full-wave simulations for long yet finite cable cases, while requiring significantly less computation time. This approach provides clear quantitative insight into the relationship between structural parameters and field penetration, facilitating the efficient design and optimization of low-frequency magnetic shielding structures.

Our research investigates a variable magnetic flux motor that utilizes magnetostrictive materials where the relative permeability changes in response to the stress induced by centrifugal force. To realize this concept, development of materials with high magnetostriction is essential. We prepared materials exhibiting large magnetostriction using iron-based binary alloys and investigated their underlying magnetostrictive mechanisms. Alloy ingots with nominal compositions of Fe100−xMx (x = 0 to 20; M = Si, Al, Ge, Ga, Ni, Co, and V) were prepared by arc melting. High magnetostriction was obtained in Fe80Ga20 (58 ppm) and Fe80Al20 (47 ppm). The magnetostriction of iron alloys is determined by the balance between the repulsive force from the anisotropy of the electron density distribution on the 3d electron orbital (corresponding to coercivity Hc) and the attractive force from the metallic bonding strength among iron atoms (corresponding to the lattice constant a).

Spiral acoustic structures hold significant potential for noise suppression in fluid flow systems; however, their aerodynamic performance has often been overlooked in previous studies. To bridge this gap, the present study develops a sequential optimization strategy aimed at simultaneously maximizing acoustic transmission loss (TL) and minimizing flow drag (FD). The strategy integrates a multi-objective genetic algorithm with surrogate models for initial global exploration of the design space, followed by an adjoint optimization method for local refinement of the inner wall geometry. Under fixed outer dimensions, the spiral geometry is parameterized using inner diameter, spiral angular frequency, spiral position, and dimensionless pitch. Sensitivity analysis is employed to identify the key structural parameters. Surrogate models are constructed to efficiently map the structural parameters to FD and total transmission loss (TY), respectively. A multi-objective genetic algorithm identifies a multi-objective optimization structure achieving TY higher than 7 dB and reducing FD by 306.35 Pa at 10 m/s compared to the original structure. The adjoint optimization, building upon this multi-objective optimization structure result, further reduces the FD by an additional 89.33 Pa without compromising the TL. The sequential optimization strategy ultimately delivers a 23% increase in TY and a 53.6% reduction in FD relative to the original structure. Furthermore, the optimization results are validated through impedance tube experiments. The sequential optimization strategy provides an efficient solution for applications demanding both effective noise control and low flow drag.

Deep neural networks are sensitive to input perturbations, such as lighting changes, undermining reliability in real-world applications. This paper introduces interval-analysis enhanced MobileNet, which is a robust image classification framework that integrates interval bound propagation into the lightweight MobileNet architecture. By modeling intensity variations as bounded intervals, the method computes guaranteed output bounds that ensure prediction stability across all perturbed inputs within the defined range. The approach reveals high accuracy without any problem in computational efficiency, making it suitable for edge deployment. The network has been trained using a hybrid loss that jointly optimizes standard accuracy and robustness margins under interval uncertainty. The experiments on CIFAR-10 and ImageNet-100 demonstrate extreme performance improvements across all intensity perturbations based on natural accuracy, far exceeding adversarial training, data augmentation, and randomized smoothing. The results show that IA-MobileNet achieves a high level of accuracy while also providing formal robustness. This bridges the gap between efficiency and reliability in vision systems that perform under unpredictable environmental conditions.

The demand for high-resolution ultrasound imaging is increasing, particularly in medical diagnostics, where the number of measurements is often limited due to physical and technical constraints. Ultrasonic tomography, based on linearizing the inverse acoustic scattering problem, enables the detection of small variations in a reference medium. To address these challenges, this study proposes a novel combination of resolution-jumping and beamforming techniques. The resolution-jumping approach progressively refines reconstruction accuracy by leveraging coarse-to-fine information, improving performance under limited measurement conditions. Simultaneously, beamforming employs multiple transmitting elements to generate a narrow beam, effectively reducing noise and enhancing structural detail recovery. Numerical simulations demonstrate that the proposed method significantly improves reconstruction quality while considerably reducing processing time, making it well-suited for real-time, high-resolution ultrasound imaging applications.