Search Results (4,149)

Smart energy meters are widely deployed in modern distribution networks, extending their role beyond revenue billing to real-time monitoring and data-driven smart city applications. However, conventional legal metrology frameworks rely on periodic recalibration and are not intended for the detection of accuracy drift or unexpected malfunctions between scheduled inspections. In scientific publications, various techniques for remote smart meters’ error surveillance are presented, but experimental verification on real distribution network data remains limited. The objective of this study is to experimentally verify two previously proposed power event-driven methods for remote estimation of active power measurement error in individual consumer meters, using a feeder-level sum meter as a reference instrument. One-second resolution electrical readings were collected from a real low-voltage distribution branch using ESP32-based local adapters communicating via MQTT over Wi-Fi, with SNTP-based clock synchronization for power event correlation. Under optimized detection parameters, the linear regression method achieved 0.20% RMSE and 0.75% maximum absolute error, and the neural network method 0.09% RMSE and 0.31%, confirming suitability for Class 1 m accuracy surveillance. Feasibility analysis of three MQTT-based deployment scenarios demonstrates that binary encoding limits local adapter buffers to 2.8 kB and worst-case daily channel demand to 2000 kB, confirming the practical viability of the proposed architecture. Full article

In complex urban environments, GNSS satellite signals suffer from severe multipath errors caused by building occlusion and reflection, which significantly degrades the accuracy of precise point positioning (PPP). This paper proposes a deep-learning-based multipath hemispherical grid correction model (DL-MHM) that integrates combined filtering and satellite embedding mechanisms. The model adopts the multi-system interoperable MHM framework to achieve effective multipath error correction. First, pseudorange and carrier phase observation residuals are calculated using the ionosphere-free combination for PPP. Then, a joint median and Kalman filtering scheme is applied to suppress noise in multi-day continuous residual sequences. A transformer-based time-series learning model is constructed, which introduces satellite-specific embedding vectors to characterize the differences between individual satellites and deeply fuse temporal features. This enables the model to adaptively fit the residual variation patterns of different satellites and accurately extract multipath errors. Finally, the multipath components predicted by the deep learning model are incorporated into the multi-system interoperable MHM model to generate the final multipath corrections. Test results show that in heavily obstructed urban scenarios, the root mean square (RMS) values of the east (E), north (N), and up (U) coordinate residuals are improved by 49.27%, 1.80%, and 3.35%, respectively, after DL-MHM correction compared to the uncorrected data. In open-sky environments, the corresponding improvements are 7.70%, 5.48%, and 34.28%. In all experimental scenarios, the proposed method outperforms both the conventional multipath hemispherical map (MHM) model and the convolutional neural network-long short-term memory (CNN-LSTM)-based MHM model in terms of overall multipath correction performance. The experimental results demonstrate that the proposed DL-MHM model can effectively mitigate multipath errors in complex urban scenarios and significantly improve the accuracy of GNSS precise positioning. Full article

Accurate representation of the pressure drop–flow rate (Δp–q) relationship of nozzle-type inflow control devices (ICDs) is critical for reliable reservoir–wellbore coupled simulation. Conventional ICD models in reservoir simulators rely primarily on empirical correlations or tabulated data, but commonly used formulations cannot consistently capture the linear behavior in the low-flow regime or the transition between flow regimes, which may reduce physical fidelity and numerical robustness. To overcome this limitation, this study proposes a unified characteristic-curve representation that integrates linear, transitional, and quadratic flow regimes into a single continuous and differentiable function through a physically constrained least-squares formulation, and further develops a physics-informed neural network (PINN) to learn the ICD pressure–flow relationship while enforcing physical consistency. The trained PINN model is embedded into a multi-segment well model within a reservoir–wellbore coupled simulation framework and evaluated using a mechanistic reservoir model containing permeability streaks with varying permeabilities. The results show that the proposed method improves numerical convergence and accurately reproduces ICD pressure–flow behavior across multiple flow regimes, providing a more physically consistent and robust representation of ICD performance for inflow control analysis and reservoir simulation. Full article

Online parameter identification is important for sensorless permanent magnet synchronous motor (PMSM) drives because motor parameter variation can reduce the accuracy of the controller and observer. However, in the background of sensorless control, the accuracy of online parameter identification is significantly affected by rotor position estimation errors and inverter nonlinearity. To address these problems, this paper proposes a high-frequency d-axis voltage injection-based online parameter identification method with inverter nonlinearity compensation. The proposed online identification method can identify the stator resistance and d-axis inductance independently. It not only overcomes the rank-deficiency problem in conventional voltage-equation-based identification, but also shows through theoretical analysis that the identification results are insensitive to rotor position estimation errors. To improve the identification accuracy, the influence and importance of inverter nonlinearity on parameter identification are analyzed, and a compensation method based on zero-sequence voltage characteristics and a feedforward neural network is developed. The identified voltage error is compensated through equivalent dead-time correction. Simulation and experimental results verify the advantages of the proposed method under different operating conditions. Full article

Sequential data prediction is a crucial area in healthcare. Healthcare data have the characteristics of non-stationarity, long-range dependence (LRD), and irregular sampling. Modeling these complex temporal features is highly challenging. Recurrent Neural Networks (RNNs) and their variants are limited in learning long-range dependencies (LRDs) due to the inherent issues of vanishing and exploding gradients. Transformers alleviate this limitation by using the self-attention mechanism. Its quadratic computational complexity and memory bottleneck limit its scalability in long-range healthcare data. In this context, Structured State Space Models (SSMs) have emerged as a promising alternative. Compared with conventional RNNs, they can alleviate the difficulty of modeling LRDs more efficiently, and many modern SSM variants achieve linear time sequence modeling while reducing the computational burden associated with Transformers. In this review, we provide a formal definition of Healthcare Process Modeling, compare the core theoretical frameworks of RNNs, Transformers, and SSMs, trace the architectural evolution of SSM architectures, and provide a comprehensive review of healthcare applications and open challenges, including LSSL, S4, S5, Mamba, and their related variants. Existing studies suggest that structured SSMs are promising for selected long-sequence healthcare prediction tasks, particularly when computational efficiency and long-context retention are important. With these advantages, they may help alleviate the computational burden in certain healthcare tasks and provide a basis for further exploring the practical application of data-driven healthcare systems in clinical practice. Full article

The sign of the maximal almost sure Lyapunov exponent determines the stability of stochastic systems, while its numerical computation for three-dimensional linear Stratonovich stochastic differential equations remains challenging due to the failure of classical two-dimensional strategies. The spherical angular motion of 3D systems produces a Fokker–Planck equation with intractable mixed partial derivatives, preventing conventional analytical solutions. This paper develops a unified computational framework for three-dimensional linear Stratonovich stochastic systems using analytical derivation for degenerate cases and physics-informed neural network (PINN) approximation for general non-degenerate scenarios. For degenerate systems, we reduce the coefficient matrix to a lower triangular form via orthogonal transformation and establish tight upper bounds based on the logarithmic growth property of the Wiener process, yielding closed-form expressions for the maximal almost sure Lyapunov exponent under all parameter sign configurations. For non-degenerate systems, we reformulate the Fokker–Planck equation in spherical coordinates and construct a customized PINN with trigonometric encoding to enforce periodic boundary conditions. The network is trained by joint loss functions of equation residuals, boundary constraints and normalization consistency, and the converged stationary density is substituted into the Furstenberg–Khasminskii formula to calculate the exponent via Gauss–Legendre quadrature. Monte Carlo simulations confirm the accuracy and robustness of the proposed method, which reliably identifies the sign of the maximal almost sure Lyapunov exponent even in near-critical regimes. Numerical experiments on a 3D stochastic Hopf bifurcation model show that noise negatively shifts the bifurcation point, with the offset linearly proportional to the squared noise intensity. This work extends Lyapunov stability analysis from two-dimensional to three-dimensional linear Stratonovich stochastic systems, offering an effective tool for stability evaluation of general three-dimensional stochastic dynamical models. Full article

Background: Viral vectors are indispensable tools in gene therapy and neural circuit mapping, offering promising therapeutic strategies for diverse genetic diseases and advancing neuroscience research. To achieve high transduction efficiency while mitigating impurity-induced immunogenicity, the development of viral vectors with improved purity and quality is essential. However, this critical requirement is often unmet by conventional purification methods such as ultracentrifugation, which are time-consuming and frequently result in limited product purity. The pseudorabies virus (PRV) is extensively employed as a viral tool for mapping neural circuits, where improved purity contributes to enhanced accuracy of neural tracing. PRV531 is a retrograde trans-synaptic tracer modified from the PRV Bartha strain, specifically designed to facilitate the precise visualization of hierarchical neural networks. Methods: In this study, we developed a method for the concentration and purification of PRV531 by integrating hollow fiber ultrafiltration (HF) with Capto^TM Core 700 (CC700) chromatography. Initially, to concentrate the viral supernatant, a 500 kDa HF membrane was employed, maintaining a feed flow rate of 80 mL/min, a shear rate ranging from 2000 to 6000 s⁻¹, and a transmembrane pressure (TMP) between 0.5 and 1 bar. Following concentration, the virus underwent purification through CC700 chromatography, operating at linear flow rates ranging from 100 to 300 cm/h. Results: Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) revealed distinct bands consistent with the expected sizes of major PRV structural proteins, each with molecular weights ranging from 25 kDa to 150 kDa, concurrently demonstrating a substantial reduction in host cell proteins (HCPs) contamination. The purified PRV531 achieved a high final infectious titer of 3.55 × 10⁹ PFU/mL, with an overall functional virus recovery of 8.88% from the crude supernatant to the final product. Conclusion: These data demonstrate that TFF combined with CC700 resin can efficiently purify retrograde trans-synaptic PRV tracer. Furthermore, this approach provides a promising strategy for purifying other viral-based tracers that traditionally rely on conventional centrifugation methods. Full article

Spaceborne neural processing units (NPUs) increasingly support real-time deep-learning inference, but their dense multiply-accumulate arrays are vulnerable to radiation-induced soft errors. Conventional radiation-hardening methods improve reliability through hardware redundancy, but they incur substantial area, performance and compiler-mapping overheads. This paper proposes tile-resilient algorithm-based fault tolerance (TR-ABFT), a software-scheduled, detection-oriented scheme for quantized NPU inference. TR-ABFT generates checksum information at tile granularity and maps checking tasks onto the original processing element (PE) array without changing the hardware topology. To make ABFT compatible with INT8 datapaths, we design two checksum-coding strategies: checksum decomposition and modulo-239 checksum coding. The modulo-239 scheme removes structural missed detections for two-bit flips with bit-position spacings in (1, 31), while preserving compatibility with signed INT8 inputs. Evaluations on ResNet, YOLOv8, and RT-DETR show that, on a $16\times 16$ array, TR-ABFT introduces only 6.37% to 24.61% additional computational overhead. By converting spatial redundancy into schedulable temporal redundancy, TR-ABFT preserves systolic-array regularity and provides a low-overhead reliability-enhancement mechanism for space-grade neural-network accelerators. Full article

Low-frequency sound insulation remains a major challenge for conventional passive materials, as improved attenuation is usually achieved at the expense of increased thickness and mass. In this work, a smooth fixed third-order ring-curve fractal-maze acoustic metamaterial is proposed for compact low-frequency sound insulation, and a physics-guided long short-term memory–physics prediction surrogate network (LSTM–PPS-Net) tandem framework is developed for its full-spectrum inverse design. Different from conventional Hilbert-type, right-angled, or sharply folded labyrinthine structures, the proposed topology uses recursively arranged curved channels to extend the effective acoustic propagation path and enhance phase accumulation within a limited space. Based on this mechanism, four physically meaningful parameters, namely slit width d, characteristic radius R₃, wall thickness t_w, and inter-column spacing l_E, are selected to construct a low-dimensional design space. A COMSOL–MATLAB automated finite-element method (FEM) workflow is established to generate 1000 valid transmission-loss (TL) spectra over 100–1700 Hz with a 5 Hz interval. For forward prediction, PPS-Net is developed by integrating geometry encoding, frequency-conditioned spectral decoding, and peak-weighted learning. The proposed PPS-Net achieves the best prediction accuracy among the tested models, with a mean absolute error (MAE) of 0.75 dB, a root mean square error (RMSE) of 1.88 dB, and a coefficient of determination (R²) of 0.96, outperforming multi-layer perceptron (MLP), convolutional neural network (CNN) and Transformer models under the same dataset and training protocol. For inverse design, the LSTM encoder extracts frequency-ordered spectral features from the target TL curve, while the frozen PPS-Net decoder provides differentiable acoustic-response feedback, thereby addressing the non-unique mapping from acoustic response to structural parameters. Furthermore, a compactness-oriented optimization strategy is introduced to balance spectral consistency, peak alignment, bandwidth preservation, and occupied-area reduction. In two representative cases, the optimized designs reduce the occupied area by approximately 21% in both representative cases, while maintaining the target attenuation characteristics after FEM verification. These results demonstrate that the proposed framework provides an efficient and physically interpretable route for the full-spectrum inverse design and compact optimization of low-frequency acoustic metamaterials. Full article

Extrapyramidal symptoms (EPS) are motor side effects commonly induced by antipsychotic medications and can lead to measurable changes in handwriting patterns. These symptoms affect both the spatial and temporal characteristics of writing, including stroke thickness, direction and the rate of directional change. To model these complex variations, we propose a novel Liquid Generative Adversarial Network (LiquidGAN), which combines the adaptive dynamics of liquid neural networks with the data generation capability of GANs. Handwriting data were collected from 94 patients with confirmed EPS and 30 healthy controls using Archimedean spiral patterns drawn with both hands. A total of 211 images were processed for both binary and multiclass classification using a pretrained ResNet50 model. The pretrained ResNet50 achieved 92% accuracy and 97% precision in the binary classification task; however, its performance dropped significantly to 57% accuracy in multiclass classification, indicating limited capability in capturing fine-grained EPS severity variations. In contrast, the proposed LiquidGAN demonstrated excellent performance in the binary classification task, achieving 97% accuracy and 98% precision. More importantly, LiquidGAN substantially outperformed the baseline in the more challenging multiclass setting, achieving 70% accuracy and precision across four classes (mild, moderate, severe, and control). This shows that the diverse dataset from the liquidGAN significantly improves the HOG-ANN classification and effectively captures complex and subtle handwriting variations associated with different EPS severity levels that conventional models such as ResNet50 fail to distinguish. In addition, LiquidGAN generated diverse and realistic synthetic handwriting samples, yielding improved Fréchet Inception Distance (FID), precision, and recall compared with style GAN. These findings demonstrate that handwriting biomarkers, when analyzed through dynamic generative learning, offer an effective and non-invasive approach for monitoring extrapyramidal side effects in clinical settings. Full article

The rapid growth of Deep Neural Networks (DNNs) has led to the development of application-specific DNN accelerators. Conventional 2D von Neumann architectures suffer from memory bandwidth limitations between the memory and the processing core. 3D DNN accelerators have emerged as a promising solution by leveraging 3D integration to enable near-memory logic or in-memory computation. By shifting computation closer to memory, these accelerators significantly reduce data movement and therefore latency, resulting in more energy-efficient operations. Monolithic 3D (M3D) integration, in particular, enables high-bandwidth systems by utilizing high-density monolithic inter-tier vias (MIVs). This paper provides a critical review of recent advances in 3D DNN accelerators that combine near-memory and compute-in-memory with various 3D technologies, offering a useful discussion and future prospects of the available technologies and architectures that have advanced the performance of DNN accelerators. Particular attention is devoted to accelerators for emerging transformer-based large language model (LLM) networks due to the higher memory demands. Thermal-aware design techniques of 3D DNN accelerators are also discussed as a means to address the fundamental challenge of heat dissipation. A detailed review is finally conducted on package-level constraints, considering signal integrity, power delivery, and thermo-mechanical reliability. Full article

To address the uneven distribution of secondary-suspension loads, the insufficient prediction accuracy of conventional mechanistic models, and the limited comprehensive optimization capability of existing leveling algorithms in the weighing and leveling process of high-speed railway carbodies, this study proposes a secondary-suspension load distribution optimization method that integrates radial basis function (RBF) neural-network error compensation with a multi-objective improved whale migration optimization algorithm. First, a mechanistic model describing the relationship between secondary-suspension loads and shim thickness is established based on the four-point weighing mechanics, and an RBF neural network is employed to compensate for the model prediction error, thereby improving the characterization accuracy of the actual loading state of the carbody. Second, a multi-objective optimization model for carbody weighing and leveling is formulated by taking the load standard deviation, total shim thickness, and number of shim positions as optimization objectives. Furthermore, the whale migration algorithm is improved according to the requirements of secondary-suspension load optimization, enabling collaborative multi-objective optimization of load deviation and total shim thickness. Finally, simulated-carbody tests and field tests on actual carbodies are carried out using a weighing and leveling test bench. The results show that the proposed method can identify the stress-free state of the carbody more accurately, reduce the error between theoretical calculations and measured data, and effectively improve the nonuniform distribution of secondary-suspension loads, demonstrating good engineering applicability. Full article

The durability of reinforced concrete is closely related to the transport behavior of water and aggressive ions within the complex nanoporous network of calcium silicate hydrate. While molecular dynamics simulations provide critical atomistic insights into these confined transport behaviors, their immense computational cost limits their scalability to complex structural and temporal domains. To overcome this bottleneck, we propose a novel, modular computational framework that synergizes high-throughput molecular dynamics with advanced graph neural networks. By rigorously learning the mapping between the local atomic environment and kinetic behaviors, our model achieves high-fidelity predictions of pore water diffusion coefficients in saturated calcium silicate hydrate while improving computational efficiency by three orders of magnitude compared to conventional force field methods. Furthermore, the model demonstrates strong transferability and can accurately capture localized nonlinear diffusion characteristics in multiparticle pore structures with rough surfaces. Building on the interchangeability of this framework’s core modules, we envision a visionary multiscale computational strategy that dynamically couples nanoscale atomistic predictions with mesoscale simulations. This work not only provides an ultrafast, highly accurate tool for screening transport properties across diverse structural configurations but also lays the groundwork for next-generation multiscale modeling of chloride ingress, ultimately advancing the design of resilient and sustainable reinforced concrete. Full article

The convergence of artificial intelligence (AI), explainable AI (XAI), and neuroscience is fostering new opportunities for understanding both machine and biological intelligence through interpretable and human-centered learning paradigms. In this Perspective, we introduce XAI2Brain as a conceptual framework for brain–AI alignment, positioning mechanistic interpretability as an intermediate layer connecting neural network representations, human understanding, and neuroscience-inspired AI design. Rather than viewing XAI solely as a post hoc transparency tool, we emphasize its emerging role in enabling mechanistic analysis of internal model representations, concept-level reasoning, and interactive human–AI alignment. We define XAI2Brain as a multi-level conceptual framework rather than a deployable system, explicitly aimed at structuring brain–AI alignment across representation-level, mechanism-level, and interaction-level perspectives. We survey the evolution of XAI methodologies—from feature attribution and concept-based explanations to mechanistic and human-centric interpretability approaches—and discuss how these methods may support bidirectional knowledge transfer between AI systems and cognitive neuroscience. Importantly, we adopt a cautious stance on brain–AI analogy, explicitly recognizing that artificial neural representations are not equivalent to biological neural representations, and instead focusing on functional and informational correspondences rather than structural equivalence. Unlike conventional human-in-the-loop or reinforcement learning from human feedback paradigms that primarily optimize behavioral outputs, XAI2Brain focuses on cognitively interpretable and mechanistically grounded alignment between AI systems and human reasoning processes. This alignment promotes interactive human-in-the-loop intelligence, empowering humans to comprehend, guide, and refine AI systems, while enabling AI systems to better interpret human instructions, intentions, and contextual reasoning. We further discuss the challenges of scaling explainability to large generative and multimodal models, including issues of interpretability robustness, cognitive compatibility, evaluation, and ethical accountability. We also highlight key limitations of current mechanistic interpretability methods, including explanation instability, representation superposition, and lack of causal guarantees, underscoring that these challenges remain open research problems. Rather than proposing a complete artificial brain architecture, this Perspective outlines a research roadmap toward more interpretable, adaptive, and neuroscience-inspired AI systems capable of supporting future brain–AI integration and collaborative intelligence. We additionally clarify that this work follows a narrative perspective review methodology with structured thematic synthesis of the literature. By framing explainability as a bridge between mechanistic AI understanding, cognitive science, and human-centered interaction, XAI2Brain highlights the importance of interpretable alignment for the next generation of brain-inspired AI systems. Full article

Wavelength stability of distributed feedback (DFB) semiconductor lasers in dense wavelength division multiplexing (DWDM) systems hinges on sub-millikelvin temperature regulation, a task complicated by the nonlinear, multi-node dynamics of the thermoelectric cooler (TEC) and the purely reactive nature of conventional proportional–integral–derivative (PID) control. We present a physics-informed neural network (PINN) built around a residual correction architecture for hybrid feedforward–feedback TEC temperature control. Rather than penalizing physics-residual violations in the loss function, the architecture wires a simplified one-node thermal model directly into the network graph as a frozen baseline. A trainable branch then learns only the residual mismatch. Temporal lag features are appended to the input so that the network can reconstruct unmeasured internal thermal states from the cold-side temperature history, which proves essential for overcoming the partial-observability bottleneck inherent in multi-node TEC packages. Ablation experiments on a high-fidelity three-node TEC simulator show that all model variants (PINN, physics-feature-augmented NN, and pure NN) exceed R² = 0.993 when trained on the full dataset, yet the PINN’s advantage becomes pronounced under data scarcity. At a 3% training budget, it reaches R² = 0.966 versus 0.930 for the pure NN, implying an approximately 5.4× reduction in the data needed to reach a given accuracy target. In closed-loop validation, the PINN+PID hybrid settles 60% faster than standalone PID. Tracking RMSE drops by 69%, and peak disturbance deviation falls by 74%, across step, multi-setpoint, and current-perturbation scenarios. All results reported here are obtained in simulations. Experimental validation on physical DFB-TEC hardware is left to future work. Full article