Search Results (75)

The rapid expansion of the Internet of Things (IoT) has increased the risk of large-scale Distributed Denial-of-Service (DDoS) attacks. In high-availability IoT environments, the operational costs of false positives and false negatives are asymmetric, whereas conventional deep learning models usually optimize static accuracy-based objectives. To address this, we propose CDRL-QNN, a cost-aware and chaos-driven reinforcement learning quantum neural network framework in which a parameterized quantum circuit serves as the action-value function approximator within a Deep Q-Network (DQN) agent. The framework incorporates asymmetric operational penalties through both the reward function and sample-wise weighted Bellman optimization, while a logistic-map-based deterministic perturbation mechanism is used to promote exploration under constrained quantum-circuit training conditions. Evaluated on a computationally constrained balanced subset of the CIC-DDoS2019 dataset, the proposed framework reduced false negatives from 49 to 33 without increasing false positives, improving recall from 0.9673 to 0.9780 and F1-score from 0.9738 to 0.9793 while lowering operational cost. These findings suggest that hybrid quantum representations can be integrated into cost-sensitive reinforcement learning pipelines for IoT intrusion detection under constrained experimental conditions. Full article

To address non-complete data issues such as missing, distorted, redundant and chaotic data in distribution network condition monitoring data easily caused by ship faults (e.g., sensor failure, cable short circuit, etc.), an improved optimization algorithm (IPS-SH5N1) considering the demand for incomplete information reconstruction in the distribution network of ship integrated power systems is proposed based on the SH5N1 meta-heuristic optimization algorithm. This algorithm is used to optimize the weights and thresholds of a BPNN, achieving accurate completion of non-complete information in the distribution network. The IPS-SH5N1-BPNN is applied to conduct information reconstruction verification on 2000 groups of incomplete data samples of ship integrated power systems. The results show that the information reconstruction accuracy (MSE) of this method is improved by more than 90% compared with the BPNN; the model training convergence time and global optimization time consumption are significantly shortened, and the average online single information reconstruction time is reduced to 4.45 ms. This method has core advantages of high precision, fast convergence, strong robustness and excellent real-time performance, which can provide reliable technical support for the intelligent operation and maintenance of ship integrated power systems. Full article

Background: Middle-aged women with obesity frequently exhibit postural misalignment and impaired balance control, which may increase the risk of functional limitations and falls. This study aimed to compare the effects of Pilates circuit training and weight circuit training on body composition, pelvic alignment indices, and balance performance in obese middle-aged women. Methods: Eighteen women (body fat ≥ 30%) were randomized to either a Pilates circuit training group (PCG, n = 9) or a weight circuit training group (WCG, n = 9) in an exploratory comparative study. Both groups performed supervised exercise three times per week for eight weeks. Outcome measures included body composition, pelvic alignment indices, dynamic balance (Y-Balance Test), and static balance (BESS). Data were analyzed using a two-way mixed ANOVA to examine time, group, and interaction effects. Results: Both groups showed significant reductions in body weight (PCG: −3.09 kg; WCG: −2.00 kg), percentage body fat (PCG: −1.85%; WCG: −1.53%), and waist-to-hip ratio (PCG: −0.05; WCG: −0.04) (p < 0.01). Significant improvements in pelvic alignment indices were observed primarily in the PCG, whereas the WCG showed smaller changes. Dynamic and static balance improved in both groups, with greater improvements observed in the PCG. Conclusions: Both training modalities improved body composition and balance outcomes in obese middle-aged women. Pilates circuit training may be associated with greater improvements in pelvic alignment and balance; however, these findings should be interpreted cautiously due to the exploratory design and small sample size. Further adequately powered randomized controlled trials are required to confirm these findings. Full article

Large-scale Virtual Power Plants (VPPs) are increasingly essential as Distributed Energy Resources (DERs) assume ancillary service duties once supplied by conventional generation, yet scaling a VPP exposes a persistent trilemma among economic efficiency, data privacy, and operational security. Centralized coordination can approach optimal revenue but requires collecting fine-grained DER operational data and creates a single point of compromise. Federated Learning (FL) mitigates raw data centralization by keeping measurements and experience local, but it introduces a fragile trust assumption that the aggregator will correctly and fairly combine model updates. This trust gap is acute in reinforcement learning-based VPP control because aggregation deviations, including selectively dropping updates, manipulating weights, replaying stale models, or injecting a replacement model, can silently bias the learned policy and degrade both profit and compliance. We propose a zero-knowledge federated reinforcement learning framework for trustless VPP coordination in which each DER trains a local deep reinforcement learning agent to solve a multi-objective dispatch problem that balances ancillary service revenue against battery degradation under operational and grid constraints, while the global aggregation step is made externally verifiable. In each round, participants bind membership via signed receipts and commit to their updates, and the aggregator produces a zk-SNARK, proving that the published global parameters equal the agreed aggregation rule applied to the receipt-bound set of committed updates under a fixed-point encoding with range constraints. Verification is lightweight and can be performed independently by each DER, removing the need to trust the aggregator for aggregation integrity without centralizing raw DER operational data or trajectories. The proposed design does not aim to hide model updates from the aggregator. Instead, it provides external verifiability of the aggregation computation while keeping raw measurements and local experience. We formalize the threat model and verifiable security properties for aggregation correctness and update inclusion, present a circuit construction with proof complexity characterized by model dimension and fleet size, and evaluate the approach in power and cyber co-simulation on the IEEE 33 bus feeder with ancillary service signals. Results show near-centralized economic performance under benign conditions and improved robustness to aggregator side deviations compared to standard federated reinforcement learning. Full article

In offshore wind farms, modular multilevel converters (MMCs) may operate under a deloading condition to accommodate wind-speed volatility and dispatch constraints. Here, deloading is defined as transmitted power < 0.2 pu (scenario S2, low-power non-reversal). Under this condition, submodule capacitor-voltage fault signatures are weak and exhibit strong operating-point-dependent drift, which degrades conventional threshold-based or offline-trained methods. We propose a lightweight switch-level IGBT open-circuit fault localization framework for deloaded MMCs. Wavelet packet decomposition is used to extract time–frequency energy features, and principal component analysis reduces feature dimensionality for lightweight deployment. An enhanced XGBoost model further integrates severity-index weighting to alleviate class imbalance and incremental learning to adapt to condition drift induced by wind-power fluctuations. MATLAB2024b/Simulink results show 99.6% accuracy in S2 with less than 2 ms inference latency, and robust performance in extended scenarios including partial-power operation and power reversal. Full article

Background: Previous research suggests that resistance training in hypoxia can cause physiological and muscle adaptations. However, this method may not be efficient for individuals who are training to optimize maximal strength and power. Objective: This study aimed to investigate the effects of 8 weeks of high-intensity resistance circuit in normobaric hypoxic conditions on maximal and explosive measures of muscle strength in upper and lower limbs. Methods: A total of 28 subjects were randomly assigned to either hypoxia (fraction of inspired oxygen [F_IO₂] = 15%; HRC_hyp: n = 15; age: 24.6 ± 6.8 years; height: 177.4 ± 5.9 cm; weight: 74.9 ± 11.5 kg) or normoxia [F_IO₂] = 20.9%; HRC_norm: n = 13; age: 23.2 ± 5.2 years; height: 173.4 ± 6.2 cm; weight: 69.4 ± 7.4 kg) groups. Training sessions consisted of two blocks of three exercises and the training intensity was fixed performed at six repetition maximum. Participants exercised twice weekly for 8 weeks, and upper and lower body power tests were performed before and after the training program. The statistical analysis applied was a two-way analysis of variance with repeated measures and Bonferroni post hoc. Results: No significant differences were observed between groups. However, the hypoxia group showed higher intra-group differences in absolute (N) (F = 7.97; Δ7.3%; p < 0.05; ES = 0.49) and relative (N/Kg) (F = 8.34; Δ7.2%; p < 0.05; ES = 0.49) maximum push-up force after the training period. Conclusions: Hypoxic circuit training may improve a specific upper body performance outcome, but no clear advantage over normoxia was observed. Full article

This research develops a decision support system for prioritizing Battery Energy Storage System (BESS) installations at existing closed-circuit television (CCTV) camera locations experiencing power interruptions in Latkrabang subdistrict. The methodology integrates nine validated features: outage frequency, downtime duration, maximum outage duration, Net Present Value (NPV), combined ROI, outage impact score, annual BESS cost, combined risk score, and UPS installation cost, derived from historical power outage records (2020–2023) and engineering economics calculations. An unsupervised K-means clustering algorithm, validated through silhouette analysis and the elbow method, categorizes installations into five risk levels, namely critical, very high, high, medium, and low, addressing the absence of predefined ground truth labels. Subsequently, Support Vector Machine (SVM) with hyperparameter optimization classifies priority installations using stratified train-test splitting (80:20). The model was initially developed and validated using 82 CCTV cameras from Lamphla Tiew subdistrict (the pilot area). The validated model was then successfully applied to 101 CCTV cameras in Latkrabang subdistrict (the target area), identifying 27 critical installation points requiring immediate BESS deployment. The weighted recommendation system balances data-driven clustering with scoring: NPV (35%), outage impact (25%), combined ROI (20%), maximum outage duration (10%), and BESS cost efficiency (10%). Full article

In order to achieve better accuracy, modern models have become increasingly large, leading to an exponential increase in computational load, making it challenging to apply them to edge computing. Binary neural networks (BNNs) are models that quantize the filter weights and activations to 1-bit. These models are highly suitable for small chips like advanced RISC machines (ARMs), field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), system-on-chips (SoCs) and other edge computing devices. To design a model that is more friendly to edge computing devices, it is crucial to reduce the floating-point operations (FLOPs). Batch normalization (BN) is an essential tool for binary neural networks; however, when convolution layers are quantized to 1-bit, the floating-point computation cost of BN layers becomes significantly high. This paper aims to reduce the floating-point operations by removing the BN layers from the model and introducing the scaled weight standardization convolution (WS-Conv) method to avoid the significant accuracy drop caused by the absence of BN layers, and to enhance the model performance through a series of optimizations, adaptive gradient clipping (AGC) and knowledge distillation (KD). Specifically, our model maintains a competitive computational cost and accuracy, even without BN layers. Furthermore, by incorporating a series of training methods, the model’s accuracy on CIFAR-100 is 0.6% higher than the baseline model, fractional activation BNN (FracBNN), while the total computational load is only 46% of the baseline model. With unchanged binary operations (BOPs), the FLOPs are reduced to nearly zero, making it more suitable for embedded platforms like FPGAs or other edge computers. Full article

Printed circuit boards (PCBs) are critical in the electronics industry. As PCB layouts grow increasingly complex, defect detection processes often encounter challenges such as low image contrast, uneven brightness, minute defect sizes, and irregular shapes, making it difficult to achieve rapid and accurate automated inspection. To address these challenges, this paper proposes a novel object detector, YOLO-UMS, designed to enhance the accuracy and speed of PCB surface defect detection. First, a lightweight plug-and-play Unified Multi-Scale Feature Fusion Pyramid Network (UMSFPN) is proposed to process and fuse multi-scale information across different resolution layers. The UMSFPN uses a Cross-Stage Partial Multi-Scale Module (CSPMS) and an optimized fusion strategy. This approach balances the integration of fine-grained edge information from shallow layers and coarse-grained semantic details from deep layers. Second, the paper introduces a lightweight RG-ELAN module, based on the ELAN network, to enhance feature extraction for small targets in complex scenes. The RG-ELAN module uses low-cost operations to generate redundant feature maps and reduce computational complexity. Finally, the Adaptive Interaction Feature Integration (AIFI) module enriches high-level features by eliminating redundant interactions among shallow-layer features. The channel-priority convolutional attention module (CPCA), deployed in the detection head, strengthens the expressive power of small target features. The experimental results show that the new UMSFPN neck can help improve the AP50 by 3.1% and AP by 2% on the self-collected dataset PCB-M, which is better than the original PAFPN neck. Meanwhile, UMSFPN achieves excellent results across different detectors and datasets, verifying its broad applicability. Without pre-training weights, YOLO-UMS achieves an 84% AP50 on the PCB-M dataset, which is a 6.4% improvement over the baseline YOLO11. Comparing results with existing target detection algorithms shows that the algorithm exhibits good performance in terms of detection accuracy. It provides a feasible solution for efficient and accurate detection of PCB surface defects in the industry. Full article

As Battery Management Systems (BMSs) continue to expand in both scale and capacity, conventional state-of-charge (SOC) estimation methods—such as Coulomb counting and model-based observers—face increasing challenges in meeting the requirements for cell-level precision, scalability, and adaptability under aging and operating variability. To address these limitations, this study integrates a Deep Neural Network (DNN)–based estimation framework into a node-level BMS architecture, enabling edge-side computation at each individual battery cell. The proposed architecture adopts a decentralized node-level structure with distributed parameter synchronization, in which each BMS node independently performs SOC estimation using shared model parameters. Global battery characteristics are learned through offline training and subsequently synchronized to all nodes, ensuring estimation consistency across large battery arrays while avoiding centralized online computation. This design enhances system scalability and deployment flexibility, particularly in high-voltage battery strings with isolated measurement requirements. The proposed DNN framework consists of two identical functional modules: an offline training module and a real-time estimation module. The training module operates on high-performance computing platforms—such as in-vehicle microcontrollers during idle periods or charging-station servers—using historical charge–discharge data to extract and update battery characteristic parameters. These parameters are then transferred to the real-time estimation chip for adaptive SOC inference. The decentralized BMS node chip integrates preprocessing circuits, a momentum-based optimizer, a first-derivative sigmoid unit, and a weight update module. The design is implemented using the TSMC 40 nm CMOS process and verified on a Xilinx Virtex-5 FPGA. Experimental results using real BMW i3 battery data demonstrate a Root Mean Square Error (RMSE) of 1.853%, with an estimation error range of [4.324%, −4.346%]. Full article

Current object detection methods deployed in closed-circuit television (CCTV) systems experience substantial performance degradation due to domain gaps between training datasets and real-world environments. At the same time, increasing privacy concerns and stricter personal data regulations limit the reuse or distribution of source-domain data, highlighting the need for source-free learning. To address these challenges, we propose a stable and effective source-free semi-supervised domain adaptation framework based on the Mean Teacher paradigm. The method integrates three key components: (1) pseudo-label fusion, which combines predictions from weakly and strongly augmented views to generate more reliable pseudo-labels; (2) static adversarial regularization (SAR), which replaces dynamic discriminator optimization with a frozen adversarial head to provide a stable domain-invariance constraint; and (3) a time-varying exponential weighting strategy that balances the contributions of labeled and unlabeled target data throughout training. We evaluate the method on four benchmark scenarios: Cityscapes, Foggy Cityscapes, Sim10k, and a real-world CCTV dataset. The experimental results demonstrate that the proposed method improves mAP@0.5 by an average of 7.2% over existing methods and achieves a 6.8% gain in a low-label setting with only 2% labeled target data. Under challenging domain shifts such as clear-to-foggy adaptation and synthetic-to-real transfer, our method yields an average improvement of 5.4%, confirming its effectiveness and practical relevance for real-world CCTV object detection under domain shift and privacy constraints. Full article

Printed circuit board (PCB) defect detection is critical to manufacturing quality, yet tiny, low-contrast defects and limited annotations challenge conventional systems. This study develops an ECA–DCN-lite–BiFPN–CARAFE-enhanced YOLOv5 detector by modifying You Only Look Once (YOLO) version 5 (YOLOv5) with Efficient Channel Attention (ECA) for channel re-weighting, a lightweight Deformable Convolution (DCN-lite) for geometric adaptability, a Bi-Directional Feature Pyramid Network (BiFPN) for multi-scale fusion, and Content-Aware ReAssembly of FEatures (CARAFE) for content-aware upsampling. A single-cycle semi-supervised training pipeline is further introduced: a detector trained on labeled images generates high-confidence pseudo-labels for unlabeled data, and the combined set is used for retraining without ratio heuristics. Evaluated on PKU-PCB under label-scarce regimes, the full model improves supervised mean Average Precision at an Intersection-over-Union threshold of 0.5 (mAP@0.5) from 0.870 (baseline) to 0.910, and reaches 0.943 mAP@0.5 with semi-supervision, with consistent class-wise gains and faster convergence. Ablation experiments validate the contribution of each module and identify robust pseudo-label thresholds, while comparisons with recent YOLO variants show favorable accuracy–efficiency trade-offs. These findings indicate that the proposed design delivers accurate, label-efficient PCB inspection suitable for Automated Optical Inspection (AOI) in production environments. This work supports SDG 9 by enhancing intelligent manufacturing systems through reliable, high-precision AI-driven PCB inspection. Full article

Accurate evaluation of the severity of external short-circuit (ESC) faults in li-ion batteries is critical to ensuring the safety and reliability of battery systems. This study proposes a novel ESC fault assessment method based on electrochemical impedance spectroscopy (EIS) and differential feature extraction from relaxation time distributions. By comparing EIS responses before and after the short circuit, differential curves are constructed, and relevant peak descriptors are extracted to form physically interpretable feature vectors without requiring equivalent circuit modeling. Standardized feature data are further analyzed using principal component analysis (PCA) and K-Means clustering to perform unsupervised classification of fault severity. In addition, a differential evolution algorithm is employed to adaptively optimize the feature weights, enhancing the monotonic correlation between the weighted scores and actual short-circuit durations. The resulting SeverityScore provides an interpretable, mechanism-driven indicator of ESC fault severity. Experimental results demonstrate that the proposed method effectively distinguishes between mild and moderate short-circuit conditions and generalizes well across four independent battery groups. The model, trained on a single group, demonstrates strong robustness by accurately classifying the fault severity for three unseen validation groups. This data-driven framework offers a robust and model-free approach for fault evaluation, providing a promising tool for health monitoring and risk assessment in li-ion batteries. Full article

Accurate fault localization in rail transit train cables is hindered by impedance mismatch, which induces overshoot interference and attenuates reflected signals, causing traditional peak-detection methods to fail. This study proposes a novel traveling wave dual-peak identification method to address this challenge. The approach employs signal polarity normalization to eliminate phase inversion, Gaussian-weighted filtering to suppress noise and distortion, and local extrema screening to robustly isolate incident and reflected wave peaks amidst complex backgrounds including overshoot oscillations and electromagnetic crosstalk. A dual-Gaussian model is optimized via nonlinear fitting to precisely quantify peak arrival times while compensating for waveform broadening. Fault distance is derived from the optimized time difference and wave velocity. Experimental validation across single-core coaxial, twin-core coaxial, and harness cables with open/short-circuit faults at multiple distances confirms the method’s effectiveness. Results demonstrate strong linear relationships between time differences and fault distances for all cable types, with successful peak identification achieved even under severe signal attenuation or strong coupling interference. This method significantly enhances localization accuracy for rail transit cable systems under impedance mismatch conditions. Full article

Background: Previous studies showed that workplace physical activity programs (WPAPs) could improve general health among employees. However, there is a lack of correlation between oxidative redox status and the metabolic and physical fitness (PF) of workers. The objective of the study was to evaluate the improvements of a 24-week combined circuit training and mobility training program on PF, oxidative redox status, and metabolic parameters on healthy academic employees. Methods: Twenty-six university employees (52.8 ± 11.5 years) followed a 24-week WPAP composed of two circuit training sessions and one mobility training session per week. PF components were assessed through one leg stand, shoulder/neck mobility, handgrip, dynamic sit-up, jump and reach, and 2-Minute step test (2MST). Oxidative stress and antioxidant potential were evaluated through derived-Reactive Oxygen Metabolites (d-ROM) and biological antioxidant potential (BAP) tests, respectively. Metabolic measurements included total cholesterol, LDL-C, HDL-C, triglycerides, and fasting plasma glucose. All assessments were conducted at baseline and after 24 weeks. Results: D-ROM values increased significantly likely due to an acute adaptive response to exercise and a stable BAP/d-ROM ratio was maintained. At baseline, subjects with higher 2MST scores showed a better BAP/d-ROM ratio compared to those with lower 2MST scores, which was also associated with normal weight status (p < 0.05), healthy values of triglycerides (p < 0.01), and LDL-C (p < 0.01). Excluding statin-treated subjects, an intriguing shift toward a condition of enhanced antioxidant capacity was observed. Conclusions: Overall, the 24-week WPAP improved metabolic health and maintained redox balance, despite increased reactive oxygen species (ROS) production. Statin supplementation may have hidden antioxidant adaptations to physical exercise, an intriguing observation that warrants further studies. Full article