Side-scan sonar is essential to underwater target detection, yet its effectiveness is hindered by scarce annotated data and complex acoustic artifacts. This study systematically evaluates four YOLO variants, YOLOv8n, YOLOv10n, YOLOv11n, and the newly released YOLOv13n, on two public side-scan sonar datasets with limited samples and severe class imbalance. We assess detection accuracy, computational efficiency, inference speed, and transfer learning using COCO pre-trained weights, as well as the impact of optimizer choice between SGD and AdamW. The results reveal distinct strengths: YOLOv8n achieves the fastest inference at 60.98 FPS, with a competitive mAP50 of 0.906, ideal for real-time applications. YOLOv11n offers the best accuracy–efficiency balance, attaining the highest recall of 0.859 and mAP50 of 0.917. YOLOv13n demonstrates exceptional precision of 0.993 and high-IoU localization, with an mAP75 of 0.760. Transfer learning consistently boosts performance, with average mAP50:95 gains exceeding 54% on the more challenging dataset, highlighting its critical role in overcoming data scarcity. SGD generally outperforms AdamW, confirming its suitability as the default optimizer. These findings provide practical guidelines: YOLOv8 for real-time needs, YOLOv11 for balanced performance, and YOLOv13 for precision-critical tasks with ample resources. This work also establishes a benchmark for future underwater autonomous system research.

This study presents an innovative solution method for ultra-fine group slowing-down equations tailored to stochastic media with double heterogeneity (DH), focusing on advanced nuclear fuels such as fully ceramic microencapsulated (FCM) fuel and Mixed Oxide (MOX) fuel. Addressing the limitations of conventional resonance calculation methods in handling DH effects, the proposed UFGSP method (the ultra-fine group slowing-down method with the Sanchez–Pomraning method) integrates the Sanchez–Pomraning technique with the ultra-fine group transport theory to resolve spatially dependent resonance cross-sections in both matrix and particle phases. The method employs high-fidelity geometric modeling, iterative cross-section homogenization, and flux reconstruction to capture neutron self-shielding effects in stochastically distributed media. Validation across seven FCM fuel cases, four poison particle configurations (BISO/QUADRISO, Bi/Tri-structural Isotropic), and four plutonium spot problems demonstrated exceptional accuracy, with maximum deviations in effective multiplication factor k_eff and resonance cross-sections remaining within ±138 pcm and ±2.4%, respectively. Key innovations include the ability to resolve radial flux distributions within TRISO particles and address resonance interference in MOX fuel matrices. The results confirm that the UFGSP method significantly enhances computational precision for DH problems, offering a robust tool for next-generation reactor design and safety analysis.

This paper addresses the ill-posed problem in calculating subgroup parameters for resonance self-shielding within nuclear reactor physics. The conventional Pade approximation method often yields negative subgroup cross-sections lacking physical meaning due to its treatment of overdetermined nonlinear systems, making the subgroup transport equations unsolvable. To overcome this, an optimized Pade approximation method is proposed: a resonance factor criterion is used to select energy groups requiring calculation; a systematic procedure dynamically traverses background cross-section combinations starting from a minimal subgroup number, incrementally increasing it until solutions meeting accuracy constraints with positive parameters are found; and, given the insufficiency of background points, a high-resolution resonance integral table is constructed, particularly for ranges exhibiting significant cross-section variations. Numerical validation confirms the method eliminates negative parameters, ensures physical validity, and significantly improves accuracy across benchmark cases including typical fuel pins, burnt pellets, and Gd-bearing lattices. This approach effectively resolves the ill-posedness of the traditional method, offering a more robust and precise subgroup resonance treatment for high-fidelity core neutronics.

This study presents an optimized subgroup transport equation solving method to enhance the computational efficiency of resonance calculations in high-fidelity reactor core simulations. By consolidating all the resonance groups into an equivalent single group and performing fixed-source calculations only for the representative subgroup cross-sections, this method significantly reduced the computational burden compared to the conventional subgroup method. Validation studies on single-cell, 2D assembly, and 3D assembly problems demonstrated that the proposed method achieves computational accuracy comparable to the conventional approach while requiring fewer fixed-source equations. This advancement offers a promising solution for improving the efficiency of resonance calculations in high-fidelity reactor core simulations, paving the way for more accurate and computationally efficient modeling of complex reactor systems.

Resonance self-shielding in multi-resonant nuclide media is a dominant physical process in reactor neutronics analysis. This study proposes an improved subgroup method (ISM) based on Padé rational approximation, constructing a high-order rational function mapping between effective and background cross-sections to overcome the precision bottleneck of traditional DSMs and BIMs in nonlinear resonance interference scenarios. The method first generates cross-section relation data via ultra-fine group calculations, then solves subgroup parameters using a positive definite system, with a Spatial Homogenization (SPH) factor introduced for reaction rate conservation. Validation results show that ISM + SPH reduces k-infinity errors from −708 pcm (DSM) to +5 pcm for UO₂ fuel, and from −269 pcm to +45 pcm for MOX fuel with ²³⁹Pu, significantly enhancing neutron transport accuracy in complex fuel systems. This work provides a theoretically rigorous and practically applicable approach for efficient resonance modeling in advanced reactor fuel design.

This study presents an innovative solution method for ultra-fine group slowing-down equations tailored to stochastic media with double heterogeneity (DH), focusing on advanced nuclear fuels such as Fully Ceramic Microencapsulated (FCM) fuel and Mixed Oxide (MOX) fuel. Addressing the limitations of conventional resonance calculation methods in handling DH effects, the proposed UFGSP method integrates the Sanchez-Pomraning technique with ultra-fine group transport theory to resolve spatially dependent resonance cross-sections in both matrix and particle phases. The method employs high-fidelity geometric modeling, iterative cross-section homogenization, and flux reconstruction to capture neutron self-shielding effects in randomly distributed media. Validation across seven FCM fuel cases, four poison particle configurations (BISO/QUADRISO), and four plutonium spot problems demonstrates exceptional accuracy, with maximum deviations in effective multiplication factor keff and resonance cross-sections remaining within ±138 pcm and ±2.4%, respectively. Key innovations include the ability to resolve radial flux distributions within TRISO particles and address resonance interference in MOX fuel matrices. The results confirm that UFGSP significantly enhances computational precision for DH problems, offering a robust tool for next-generation reactor design and safety analysis.