Search Results (8,685)

Accurate fault location is essential for improving the reliability and service restoration capability of distribution networks. With the increasing penetration of distributed generation, power electronic devices, and flexible loads, fault transient signals become increasingly nonlinear and nonstationary, posing challenges to conventional impedance-based, traveling-wave-based, and feature-engineering-based methods. To improve transient fault feature representation, this paper proposes a wavelet-embedded residual attention convolutional neural network (CNN) for distribution network fault location. The task is formulated as a multi-class classification problem, in which each predefined line section is treated as a candidate fault location class. The proposed method embeds discrete wavelet decomposition into the convolutional feature extraction process, enabling low-frequency trend components and high-frequency transient components to be jointly represented and fused by subsequent trainable network modules. Residual connections improve deep feature propagation, and an attention mechanism enhances fault-sensitive representations. Simulation studies on the IEEE 33-bus distribution system show that the proposed method outperforms multi-layer perceptron (MLP), support vector machine (SVM), standard CNN, ResNet, and Attention-CNN, achieving 98.27% accuracy and a 98.33% F1-score. The class-wise results and robustness tests under different transition resistances, noise levels, and fault types further verify the effectiveness and adaptability of the proposed method. Full article

Direction of arrival (DoA) estimation is a fundamental problem in modern communication systems, such as 5G/6G cellular systems, V2X, and radar. Modern DoA estimation techniques enhance signal reception, mitigate interference, enhance communication efficiency, improve capacity, and improve spatial selectivity. In this paper, a two-channel residual neural network (ResNet) CNN is designed and trained based on the covariance matrix for a realistic electromagnetic antenna array model by expanding the steering vector obtained from the embedded element radiations. The regression DoA estimation is parameterized for three scenarios: regression using a trigonometric angle process, regression directly in degrees, and regression in radians. Then, the proposed network is compared with the modified conventional multiple signal classification (MUSIC), minimum variance distortion-less response (MVDR), and a two-channel deep CNN. A microstrip antenna array is designed, operating at 28 GHz, using Ansys Electronic Desktop to obtain the 3D embedded element radiation, for both co-polarized and cross-polarized components, considering mutual coupling among the antenna array elements, finite-element spacing, and array geometry. The proposed degree-based ResNet CNN achieves sub-degree azimuth and elevation RMSE for angular separations greater than 10° at an SNR of 0 dB in our simulations, clearly outperforming modified MUSIC, MVDR, and deep CNN learning-based 2D DoA methods that require significantly higher SNR to reach comparable accuracy. Moreover, the network operating directly on the real and imaginary parts of the covariance matrix and predicting angles in degrees consistently yields lower RMSE than variants trained to predict radians or sine–cosine representations, while avoiding the steering vector knowledge and postprocessing steps, spatial spectra, peak search, or root-finding, used in existing approaches. Full article

This paper presents an adaptive neural-network-based algorithm for chattering mitigation in sliding mode control (SMC) of parallel multicellular DC/AC power converters. Although conventional SMC provides strong robustness against parameter uncertainties, external disturbances, and load variations, its discontinuous control action often generates chattering, resulting in excessive switching activity and reduced converter performance. To address this limitation, a computationally efficient adaptive neural network is integrated into the SMC framework to approximate the discontinuous switching term and generate a smooth control signal. The proposed algorithm updates neural network parameters online through an adaptive learning mechanism, enabling real-time compensation of modeling uncertainties while preserving the inherent robustness of SMC. The resulting adaptive neural network sliding mode control (ANN-SMC) algorithm is formulated to ensure accurate output voltage tracking, balanced operation of converter cells, and reduced switching oscillations. Extensive simulation studies are conducted under different operating scenarios, including load variations and system disturbances. The performance of the proposed method is evaluated against classical SMC using quantitative indicators related to tracking accuracy, dynamic response, robustness, and chattering suppression. The results demonstrate that the ANN-SMC algorithm significantly reduces high-frequency oscillations while improving transient behavior and maintaining robust operation. These findings indicate that the proposed adaptive learning-based control algorithm constitutes an effective and scalable solution for advanced power conversion systems operating under uncertain conditions. Full article

A reconfigurable intelligent surface (RIS) is a promising technology for beyond-fifth-generation (B5G) and sixth-generation (6G) wireless communications, but its passive reflection and two-hop double-fading structure make cascaded channel estimation challenging. Conventional convolutional neural network (CNN) estimators require many trainable parameters, while a single shallow parameterized quantum circuit (PQC) may have limited feature representation. Deep quantum circuits can also suffer from noise and barren-plateau effects on noisy intermediate-scale quantum (NISQ) devices. To address these issues, this paper proposes a multi-filter quantum convolutional neural network (MF-QCNN) for cascaded channel estimation in RIS-assisted multi-user uplink systems. The proposed model uses multiple independent shallow PQC filters in parallel, concatenates their measured features, and estimates the cascaded channel through a compact classical dense head, with the total trainable-parameter count scaling as $182F+696$ for F parallel filters. Simulation results, compared with a single-filter quantum convolutional neural network (QCNN), CNN, and multilayer perceptron (MLP) baselines, show that at a signal-to-noise ratio (SNR) of 20 dB, the 3-filter MF-QCNN reduces the normalized mean squared error (NMSE) by approximately 22.9, 8.1, and 4.6 dB relative to the single-filter QCNN, CNN, and MLP baselines, respectively, while using only about 19.3% of the CNN trainable parameters. Under zero-forcing (ZF) precoding, it achieves the highest achievable sum rate among the learning-based estimators; at SNR = 30 dB, it improves the achievable sum rate by approximately 17.4% and 12.8% over the CNN and MLP baselines, respectively. These simulation results suggest that the parallel shallow-PQC design can serve as a compact quantum-aided estimator for RIS channel estimation and may provide a useful basis for future studies on AI-native transceiver design in B5G/6G networks. Full article

To address the low computational efficiency of conventional forward modeling that hinders real-time inversion of borehole resistivity logging data in deviated/horizontal wells through anisotropic formations, this paper presents an adaptive inversion method based on a deep neural network forward surrogate model. A resistivity response database covering deviation angle, mud invasion, and anisotropy is constructed using three-dimensional finite-element forward modeling. The deep neural network architecture is systematically optimized by varying the number of hidden layers and neurons per layer, comparing five activation functions, evaluating four training algorithms, and testing five batch size ratios. The resulting deep learning-based forward model achieves a speedup of over two orders of magnitude compared with 3D finite-element modeling while maintaining high accuracy (maximum relative error < 1%). By integrating this fast forward model with an adaptively modified Levenberg–Marquardt algorithm, rapid inversion in anisotropic formations is realized. Numerical simulations and field data processing demonstrate that the proposed method accurately extracts uninvaded resistivity, invasion depth, and anisotropy coefficient, with an efficiency gain of approximately 98% over traditional approaches. Reconstructed logs show excellent agreement with measured data, providing robust support for real-time evaluation of deviated and horizontal wells. Full article

Background/Objectives: Ankle dorsiflexion plays a fundamental role in gait stability, impact absorption, and the stance-to-swing transition, and its impairment is a major limitation in transtibial prostheses. This study proposes and evaluates a lightweight two-stage pipeline for generating ankle-dorsiflexion references using a single shank-mounted inertial measurement unit (IMU). Methods: In the first stage, a deep neural network (DNN) estimates the shank pitch waveform from raw three-axis accelerations and angular velocities. In the second stage, the estimated shank pitch is transformed into an ankle-dorsiflexion waveform using a temporal mapping model. The approach was evaluated on a multisubject subset of the NONAN GaitPrint database comprising 35 healthy young adults, 598 walking trials, and approximately 122,468 gait cycles, using a strict subject-held-out protocol. Results: A feature-based Random Forest baseline showed limited performance, whereas the waveform-based DNN achieved high accuracy for shank pitch estimation, with test R² values up to 0.97. A conventional polynomial mapping between shank pitch and dorsiflexion yielded weak performance, whereas a temporal mapping model substantially improved the estimation of ankle dorsiflexion, with test R² values up to 0.85. The resulting ankle reference was integrated into a Gazebo/Robot Operating System 2 (ROS 2) simulation of a transtibial prosthesis, where the generated trajectories were executed in a software integration test under open-loop position control, confirming stable and consistent trajectory execution. Conclusions: These results indicate that combining accurate shank pitch estimation with temporal mapping enables feasible ankle-dorsiflexion reference generation from a single sensor in able-bodied gait, offering a preliminary, simulation-based pathway for single-sensor artificial intelligence (AI) pipelines in prosthetic development. The framework supports waveform-level feasibility, not clinical readiness or functional prosthetic control. Full article

The recovery of low- and medium-temperature waste heat using Organic Rankine Cycles (ORCs) is increasingly important for improving the efficiency and sustainability of industrial and energy systems. In compact ORC turboexpanders, high specific power output and large pressure ratios often require single- or two-stage turbines operating in transonic or supersonic regimes. Under these conditions, stator blade design is complicated by strong compressible-flow effects and, for organic working fluids, by real-gas thermodynamic behavior. Conventional supersonic stator design methods, such as the method of characteristics, are mainly applicable to the diverging supersonic portion of the blade passage, while the converging region is typically defined using empirical or heuristic prescriptions. This paper presents a physics-informed neural-network-based design method for supersonic turbine stator blades. The proposed framework generates the complete inter-blade passage, including both the converging and diverging regions, starting from a prescribed mean-line geometry and Mach number distribution. The velocity field is obtained by solving the governing equations of steady, inviscid, adiabatic, irrotational compressible flow within a PINN formulation. A hard boundary-condition strategy is used to impose the specified mean-line velocity distribution exactly, while real-fluid thermodynamic effects are incorporated through lookup tables for the speed of sound and density. The blade contours are then reconstructed from stream-function isolines predicted from the computed velocity field. The method is demonstrated for two working fluids: air, treated as a perfect gas, and toluene undergoing transcritical expansion. The resulting blade passages are first validated using inviscid CFD simulations, which show close agreement between the prescribed and computed mean-line Mach number distributions. Turbulent CFD simulations of the final blade cascades confirm smooth acceleration through the inter-blade passage, with no strong internal shocks and only weak fishtail shocks downstream of the trailing edge. For both fluids, the post-expansion ratio is approximately unity and the exit flow angle remains close to the prescribed blade metal angle, indicating well-matched supersonic stator designs. The results demonstrate that the proposed PINN-based design method provides a physically consistent approach for generating supersonic stator blade profiles for both ideal-gas and real-gas turbine applications. Full article

Agent-based neural models often encode transmission within neuron state updates, which can make it difficult to separately log and quantify spatial recruitment patterns, delay structure, and environment-mediated feedback effects. We present LANA (Local Adaptive Neural Agents), a dual-agent neural agent-based model in which neurons and propagating signals are represented as distinct interacting entities embedded in a dynamic environmental field. The model combines discrete leaky integrate-and-fire neuron dynamics, mobile signal agents, synaptic links with distance-dependent delays, and a bounded environment-to-neuron feedback mechanism. LANA is intended as a normalized phenomenological mesoscopic framework for mechanism-level comparison rather than as a circuit-specific biophysical reconstruction. To support interpretability and reproducibility, we report a compact internal verification block for the implemented operators, including delay propagation, environmental decay and diffusion, threshold activation, and refractory enforcement. We then compare the full LANA model against a matched neuron-only baseline and summarize spatial recruitment using first-spike maps, cumulative recruitment times, and wavefront speed as a secondary descriptive metric. Finally, we evaluate two controlled operating regimes, a resting regime (S1) and a hyperexcitable regime (S2), under fixed network size, stimulation schedule, and matched random seeds. Relative to the baseline, the full model sustains and spreads activity more effectively and provides spatially resolved recruitment summaries, including first-spike timing and cumulative recruitment measures, that are not available in the same form when transmission is represented only through neuron-level updates. Relative to S1, S2 exhibits earlier activation, higher firing activity, stronger environmental accumulation, and faster cumulative recruitment. Local and factorial sensitivity analyses further identify the parameters that most strongly govern these regime differences. Together, these results position LANA as a normalized mesoscopic and computationally tractable framework for studying how excitability, transport state dynamics, delayed coupling, and environment-mediated feedback jointly shape emergent activity in controlled simulation settings. Full article

This paper investigates an observer-based secure control method for networked non-Lipschitz nonlinear systems subject to unknown nonlinearities, external disturbances, sensor noises, and intermittent denial-of-service (DoS) attacks. Multi-layer neural networks (MNNs) are adopted to compensate for non-smooth, non-Lipschitz terms, guaranteeing bounded approximation errors. A resilient high-gain observer fused with the MNN is developed to continuously reconstruct system states. When DoS attacks block sensor channels, the observer acts as a virtual dynamic engine to substitute for lost real-time measurements, providing uninterrupted feedback to the controller. Furthermore, to optimize communication efficiency, an observer-based static event-triggered mechanism (SETM) coupled with a hold-input strategy is integrated. Employing the Lyapunov–Krasovskii functional method, sufficient conditions are derived to prove that the closed-loop system remains uniformly ultimately bounded (UUB) under the joint effects of approximation errors, disturbances, and attacks. Simulation results on a two-link manipulator demonstrate that the proposed secure control scheme effectively counters aggressive DoS attacks while achieving a 56.8% reduction in network transmissions compared with conventional periodic sampling paradigms, striking a favorable balance between tracking accuracy and resource efficiency. Full article

Existing solutions for resource allocation in Zero Trust (ZT) SDN networks treat security, resilience, and efficiency separately; centralised approaches violate data privacy; deep reinforcement learning (DRL) lacks trust dynamics; and federated learning (FL) has not incorporated graph neural networks (GNNs) or failure resilience. These limitations motivate a federated GNN-DRL framework that preserves data locality while jointly optimising trust, resilience, and performance. Each domain trains a local GNN-DRL agent with a GNN encoder for topology awareness and a hybrid DRL module (Deep Q-Network for discrete failover actions and Soft Actor-Critic for continuous bandwidth tuning) under a stochastic trust evolution modelled by stochastic differential equations (SDEs) with a reflection mechanism to ensure well-posedness. Three pseudo-code algorithms detail client-side training (Algorithm 1), server-side q-fair aggregation (Algorithm 2), and local gradient updates (Algorithm 3). Extensive simulations on a 100-node topology divided into five domains demonstrate that: (i) under low-to-moderate failures (≤20%), trust violations remain below 1.8%, and even under severe failures (40%), violations stay at 4.2% (within the 5% ZT boundary); (ii) recovery time is reduced by 53%; (iii) throughput under failures improves by 32%; (iv) compromise resistance reaches seven nodes (vs. three centralised); (v) attack surface shrinks to four nodes (vs. 98 baseline); and (vi) lateral movement containment attains 98%. The federated framework approaches centralised performance while preserving data locality, offering a practical and secure solution for multi-domain ZT, SDN networks. Full article

Ferrofluid sealing is an important non-contact sealing technology for ultra-high vacuum (UHV) equipment, but its reliability is affected by more than pressure-bearing capacity alone. This review shows that carrier-liquid evaporation, material outgassing, thermal degradation, magnetic-field distortion, and liquid-ring instability are the main factors limiting UHV ferrofluid seals. Multiphysics simulation and parametric optimization remain the most mature tools for analyzing magnetic-field distribution, pressure resistance, temperature rise, and structural deformation. Data-driven condition identification improves failure monitoring, whereas physics-informed neural networks, topology optimization, and multi-objective optimization are still emerging methods for low-sample prediction and collaborative design. Future studies should focus on low-vapor-pressure ferrofluids, bake-out compatibility, thermal management, lifetime prediction, and integrated model–data design frameworks. Full article

Structural health monitoring (SHM) is essential for assessing the safety and serviceability of bridge structures. Identifying progressive and concurrent damage remains challenging due to the complex and continuous nature of structural deterioration. This study proposes a hybrid one-dimensional convolutional neural network and long short-term memory (1D-CNN–LSTM) framework for vibration-based damage localisation and severity estimation in reinforced concrete bridges. Operational modal analysis is applied to field-measured vibration data from an in-service bridge. A finite element model is updated using particle swarm optimisation, reducing frequency discrepancies from 7–17% to within ±3%. Progressive single-, double-, and triple-element damage scenarios are simulated through systematic stiffness degradation. The resulting modal frequency data are used to train 1D-CNN–LSTM models using Pareto front optimisation. The proposed framework achieves coefficients of determination above 0.80 with low prediction errors (MSE and MAE < 2) for single- and double-element damage scenarios. The results support the use of the proposed framework for screening-level assessment of bridge damage under controlled simulated conditions. Full article

This paper addresses the high-precision position tracking control problem for the rotary direct-drive valve (RDDV) subject to complex nonlinear dynamics and unknown external disturbances. To achieve superior transient and steady-state performance, a novel adaptive neural network preset-time control (ANNPTC) strategy is proposed. Distinct from conventional finite-time or fixed-time control schemes, the proposed ANNPTC ensures that the tracking error converges to a prescribed neighborhood of the origin within a prescribed residual set after the user-defined time $T_c$ under the admissible initial condition. Specifically, adaptive radial basis function neural networks (RBFNNs) are utilized to estimate and compensate for unmodeled dynamics and disturbances, significantly enhancing the steady-state precision of the system. The uniform ultimate boundedness of all signals in the closed-loop system and the prescribed-performance property are established via Lyapunov stability analysis. Finally, extensive simulation results on a high-fidelity RDDV model demonstrate that the proposed method yields faster response speed and higher tracking accuracy compared with benchmark controllers, thereby validating its efficacy and superiority in RDDV applications. Full article

With the continuous expansion of unmanned aerial vehicle (UAV) applications, generating near-optimal paths and achieving effective obstacle avoidance in complex environments remain highly challenging tasks. To address the problems of multi-objective path planning and obstacle detection for UAV flight missions in orchard environments, this paper proposes a novel hybrid algorithmic framework named NeuroJPS-A. The main scientific contribution is the synergistic integration of neural combinatorial optimization, 3D-JPS, and adaptive APF, enabling task-aware obstacle avoidance and closed-loop trajectory adjustment. This method introduces neural combinatorial optimization from the TSP into the 3D-JPS algorithm, optimizing the search mechanism of the traditional JPS and further shortening the UAV’s globally planned path length. In addition, this study integrates the proposed algorithm with the APF to solve the local dynamic obstacle avoidance problem. Quantitative results show that NeuroJPS-A reduces path length by 10% and the number of turns by 47.8% in 2D, and achieves a 24.9% shorter path and 22% of A*’s computation time in 3D. To verify the performance of the proposed method, comprehensive simulation experiments were conducted. The experimental results demonstrate that the NeuroJPS-A algorithm enables UAVs to quickly and effectively generate optimal planned routes, ensuring safe navigation in complex orchard environments and preventing collisions during flight missions. Full article

Anomaly detection systems deployed in critical applications require not only high accuracy but also reliable uncertainty quantification and coverage guarantees. This paper is an empirical study—rather than a contribution of new conformal-prediction machinery—of class-conditional (Mondrian) conformal prediction for anomaly detection under extreme class imbalance, characterizing where standard conformal prediction fails and how class-conditional calibration restores valid coverage. Class-conditional conformal prediction constructs prediction sets that, under exchangeability, contain the true label with user-specified confidence (e.g., 90%), enabling systems to abstain on uncertain predictions. Unlike standard conformal prediction that fails catastrophically under extreme imbalance—achieving only 52.94% anomaly coverage at a 1:345 imbalance ratio—class-conditional calibration maintains 90.59% anomaly coverage by computing quantiles separately for each class. We apply the standard softmax-based nonconformity score $s=1-f_y\left(x\right)$ within each class, ensuring valid coverage for both normal and anomalous instances with coverage gaps ranging from 0.50% to 5.18% depending on dataset characteristics. Extensive experiments on three real-world datasets (Microsoft Azure KPI, Yahoo, NAB) demonstrate that the method achieves empirical coverage within 0.06–0.33% of theoretical targets at confidence levels $\alpha \le 0.05$; on the most imbalanced benchmark (Microsoft Azure KPI at a 1:345 ratio and $\alpha =0.10$), this corresponds to a 37.65 percentage point improvement in anomaly coverage over standard conformal prediction. We restate finite-sample coverage bounds and exchangeability conditions in the binary anomaly detection setting and validate them empirically through Monte Carlo simulation. Multi-model evaluation across XGBoost, Random Forest, and Neural Networks demonstrates the model-agnostic property of the framework, while also identifying conditions (poor base-classifier discrimination, small minority calibration sets) under which coverage may be marginally violated. Comparison with alternative uncertainty quantification methods (isotonic probability calibration, Monte Carlo dropout) shows that only conformal prediction provides formal guarantees while maintaining 90.59% anomaly coverage versus 76.47% and 84.71% for alternatives. The abstention mechanism identifies 34–66% of predictions as uncertain at high confidence levels (99%), enabling safety-critical systems to defer difficult cases to human experts while preserving baseline discrimination (ROC-AUC unchanged). Full article