Search Results (339)

With the increasing penetration of renewable energy systems, multilevel inverters have been widely adopted to meet the growing demand for high-power and high-quality energy conversion. Among various multilevel topologies, cascaded H-bridge multilevel inverters (CHMIs) are particularly attractive due to their modular structure and improved output voltage quality. However, the increased number of power semiconductor devices and switching states significantly complicates fault diagnosis under practical operating conditions. Currently, most existing neural networks for fault diagnosis are manually designed based on domain expertise. This may limit their adaptability to task-specific fault patterns as well as edge-side inference performance. To reduce the dependence on manually designed diagnostic networks, an edge-oriented fault diagnosis framework based on differentiable architecture search (DARTS) is proposed to automatically design task-specific diagnostic networks. A simplified special cell search strategy is adopted to improve search efficiency and facilitate practical deployment. The searched architectures are lightweight and suitable for deployment on edge platforms. The experiments show that the proposed method achieves an average diagnostic accuracy of 99.44% on the test set under the RL load of $(7\mathrm{\Omega },6\mathrm{mH})$. Furthermore, the searched model contains only 0.2417 M trainable parameters, and edge deployment experiments on the Jetson Orin Nano platform show low-latency inference capability. Full article

To address the challenges of high feature similarity and severe noise interference in the open-circuit fault diagnosis of IGBT inverter modules under harsh marine conditions, this paper proposes an improved artificial bee colony-optimized one-dimensional convolutional neural network (IABC-1D-CNN) for robust fault diagnosis in marine ranching power systems. This study provides a MATLAB R2024a/Simulink-based feasibility validation rather than hardware or field verification. First, a photovoltaic grid-connected inverter simulation model is established to generate three-phase current signals under different operating conditions and fault states, and a sliding-window segmentation method combined with data augmentation is employed to improve sample diversity. Then, the improved artificial bee colony algorithm, incorporating differential evolution and genetic strategies, is used to globally optimize the key hyperparameters of the 1D-CNN, thereby improving convergence efficiency and model stability. Based on the optimized architecture, the proposed model enables automatic feature extraction and accurate identification of IGBT open-circuit faults under complex marine environments. Experimental results show that the proposed method achieves high diagnostic accuracy under both noise-free and noisy conditions. Under signal-to-noise ratios (SNRs) of 20 dB, 15 dB, 10 dB, and 0 dB, the diagnostic accuracies reach 99.55%, 98.86%, 97.27%, and 89.25%, respectively, consistently outperforming Baseline 1D-CNN, CNN-LSTM, and ELM. These results demonstrate that the proposed method provides a simulation-validated diagnostic framework with strong classification accuracy and noise robustness, while practical deployment requires further HIL and field-data validation. Full article

The increasing penetration of renewable energy sources in inverter-dominated microgrids introduces significant challenges for maintaining voltage and frequency stability under weak-grid and dynamically varying operating conditions. Conventional inverter control strategies, including droop control and virtual synchronous machine (VSM) methods, often exhibit limited adaptability and degraded transient performance under renewable intermittency and uncertain load variations. This paper proposes a physics-informed neural-network (PINN)-based supervisory framework for real-time grid-forming inverter control. The proposed approach embeds swing-equation dynamics, Kirchhoff-based electrical constraints, and stability-aware objectives directly into the neural-network optimization process to improve physical consistency, robustness, and operational reliability. The controller is trained offline and deployed for low-latency online inference on an NVIDIA Jetson AGX Xavier embedded platform. Simulation and hardware-in-the-loop validation results demonstrate improved transient stability, reduced frequency deviation, enhanced voltage regulation, and superior robustness compared with conventional droop, VSM, and purely data-driven neural-network controllers. The proposed framework achieved an average inference latency of approximately 0.7 ms while maintaining stable operation under renewable intermittency, load disturbances, and weak-grid conditions. The results demonstrate the potential of physics-informed machine learning for supervisory real-time control of inverter-dominated microgrids and intelligent renewable energy systems. Full article

Reliable fault detection and diagnosis (FDD) plays a key role in inverter-driven permanent magnet synchronous motor (PMSM) systems, especially in applications where operational continuity cannot be compromised. In this work, a hybrid deep learning framework is developed by combining one-dimensional convolutional neural networks (CNN), bidirectional long short-term memory networks (BiLSTM), and a multi-head self-attention mechanism. The model targets multi-class fault classification in a three-phase PMSM inverter system. Its effectiveness is evaluated on a publicly available experimental dataset consisting of 10,892 multi-sensor samples collected under nine operating conditions, including normal operation, open-circuit faults, short-circuit faults, and half-bridge overheating scenarios. To avoid temporal data leakage, a block-aware chronological splitting strategy is applied. Model hyperparameters are determined through a validation process involving 24 different configurations. The proposed CNN–BiLSTM–Attention model achieves a macro F1-score of 0.9681 ± 0.0195, accuracy of 0.9810 ± 0.0102, Matthews correlation coefficient (MCC) of 0.9757 ± 0.0130, and ROC-AUC of 0.9996 ± 0.0003 over five independent runs, achieving the highest accuracy and MCC among all evaluated models; although the Random Forest baseline attains a marginally higher macro F1 score (0.9747) by operating on temporally aggregated features without temporal modelling, the proposed model provides superior discrimination across the full confusion matrix structure alongside end-to-end temporal interpretability via SHAP. Model interpretability is provided through SHAP (SHapley Additive exPlanations) GradientExplainer analysis, revealing that temperature-related features dominate fault discrimination, particularly for over-heating conditions, while current imbalance features are critical for distinguishing open- and short-circuit faults. Full article

Forest fires occur frequently around the world due to extreme weather conditions of high temperatures and drought. Vision-based convolutional neural networks (CNNs) have greatly improved forest fire detection accuracy. However, slow inference speed severely restricts real-time deployment in actual forest scenes. Existing models generally adopt group convolution (GConv) or depthwise convolution (DWConv) to reduce computational complexity, which causes frequent memory access and result in a practical inference speed far below theoretical expectations. Therefore, we propose a novel fast neural network named FasterNetFire for forest fire detection, which introduces partial convolution (PConv) as the basic feature extraction operator to reduce redundant computation as well as memory access overhead simultaneously. FasterNetFire is composed of four cascaded stages, each stage contains several stacked FasterNet Blocks, and the core of each FasterNet Block is an inverted residual module built upon PConv. The proposed network significantly improves inference efficiency while maintaining the effectiveness of spatial feature extraction. Experiments conducted on the FD and Foggia’s fire detection dataset demonstrate that our FasterNetFire achieves an impressive inference speed of up to 290 frames per second (FPS) on graphics processing unit (GPU) platforms. Compared with current representative methods, its inference speed is 4.5× and 4× faster than that of EFDNet and DFAN. Furthermore, FasterNetFire achieves the best results among 17 state-of-the-art methods, achieving an excellent balance between detection accuracy and real-time response performance. This advantage fully verifies the high efficiency of PConv in vision forest fire detection tasks and provides a novel lightweight solution for real-time monitoring and early warning of forest fires in resource-constrained environments. Full article

There are several methods which are utilized for flooding oil process monitoring, such as the seismic methods, and the electromagnetic methods. As the gas flooding oil process is complicated, conventional methods are not capable of monitoring the gas flooding oil process accurately. This study utilizes the Ground Penetrating Radar (GPR) method to monitor the CO₂ flooding oil and water flooding oil processes, as the difference in dielectric constants and conductivity of CO₂, oil and water is utilized to infer distributions of CO₂, oil and water. Moreover, as GPR data processing is time-consuming, it is impossible to process the GPR data in real-time by a conventional method, such as the full waveform inversion method. This study utilizes U-Net neural networks to invert for the subsurface dielectric constants and conductivity distributions of CO₂, oil and water in real-time. A deep learning inversion network based on the U-Net architecture is trained to extract multi-scale features through an encoder–decoder structure, achieving an end-to-end mapping from GPR echo signals to subsurface electrical parameters. The study utilizes the gprMax forward tool to simulate the dynamic response changes in rock-electrical parameters during flooding and constructs a high-resolution training dataset of 100,000 samples. Each sample contains the relationships between a subsurface electrical parameter model and its corresponding multi-transmitter, multi-receiver GPR responses. This method was first tested by the synthetic data of oil–water flooding and oil–water–gas flooding, and then it was tested by observed data from physical core experiments. Numerical and physical core experimental results show that the method accurately inverts the electrical parameter distributions of oil, water, and gas in the sandstone model, successfully capturing the position and morphology changes in the displacement front. The average relative error of dielectric constant inversion is controlled within 8% with the error mainly from the low dielectric constant regions and the relative error of conductivity is smaller than 10%, with the error mainly concentrated in high-conductivity water regions for conductivity inversion results. The results reveal the feasibility and superiority of the neural network-based deep learning method in GPR electromagnetic inversion, providing a new method for real-time flooding monitoring and intelligent reservoir development during oil and gas flooding. Moreover, the proposed approach offers a fast inversion solution and is less affected by the initial model and noise. Full article

The direct-quadrature (dq) axis control method is a widely employed approach for off-grid and grid-connected inverters in solar photovoltaic (PV) systems that can regulate active and reactive power control. Conventional fixed-gain dq-axis PI controllers may exhibit degraded transient performance and reduced harmonic suppression capability under highly dynamic operating conditions. This article proposes an innovative control scheme of an inverter-based islanded microgrid consisting of PV generation and battery energy storage systems (BESS) that can deliver stable power sharing and robust voltage regulation even under highly dynamic operating conditions. An improved inverter control method based on an artificial neural network-based proportional integral (ANN-PI) controller is investigated to accurately control the dq-axis approach for the DC-link and voltage control loops. The suggested system was validated under MATLAB/Simulink to prove the effectiveness of the proposed controller. The achieved results indicate that the ANN-PI controller presents a high convergence speed and low overshoot with a low total harmonic distortion (THD) index of 3.9% under resistive and inductive loads, thus meeting the IEEE power quality standards. Full article

Permanent magnet synchronous motors (PMSMs) suffer performance degradation under parameter uncertainties and external load disturbances, reducing the effectiveness of conventional proportional-integral and field-oriented control (FOC) schemes. This paper presents an artificial intelligence (AI) enhanced hybrid controller that combines finite-control-set model predictive control (FCS-MPC) and active disturbance rejection control (ADRC). The FCS-MPC optimizes inverter switching states by minimizing a cost function through predicted current trajectories. Additionally, the ADRC employs an extended state observer to estimate and compensate for aggregated disturbances. A lightweight radial basis function neural network is utilized, whose centers and widths are initialized offline based on k-means clustering on representative data, while its output weights are updated online via a Lyapunov-based adaptive law. This network dynamically adjusts the MPC cost function weights and ADRC observer bandwidth according to real-time operating conditions, while enabling online identification of key motor parameters. MATLAB/Simulink R2024a simulations under step load torque conditions verify that the proposed method achieves a speed deviation within 3% of the rated value, an over 90% reduction in torque ripple compared to FOC, and a settling time of less than 5 ms. Although it incurs a moderate computational cost, the proposed controller exhibits improved tracking accuracy and enhanced robustness under simulated conditions. Consequently, the AI-enhanced MPC-ADRC strategy shows strong potential for high-performance applications, subject to future experimental validation. Full article

Accurate estimation of the internal spatial-temporal temperature distribution is crucial for the safety and performance management of lithium-ion batteries. However, traditional lumped parameter models overlook spatial gradients, while numerical methods for partial differential equations (PDEs) incur high computational costs. This paper proposes a hard constraint physics-informed neural network (HCPINN) framework for the real-time reconstruction of the axial temperature field in 18,650 cylindrical batteries. By restructuring the neural network’s solution space through distance functions, the Robin boundary conditions are strictly embedded as hard constraints, ensuring exact satisfaction of the prescribed Robin boundary conditions within the mathematical model and eliminating boundary loss terms. An electro-thermal coupled model considering the Arrhenius effect and state-of-charge (SOC) dependent internal resistance is integrated into the loss function to capture the nonlinear heat generation dynamics. Experimental validation across discharge rates from 1C to 4C demonstrates that the HCPINN achieves high estimation accuracy with a mean absolute error (MAE) below 0.34 °C. Furthermore, by leveraging the continuous differentiability of the model, this study quantifies the evolution of spatial temperature gradients and reveals the ideal heat transfer coefficients required for thermal equilibrium are inverted, providing a quantitative basis for the design of advanced battery thermal management systems (BTMS). Full article

Accurate estimation of the State of Health (SOH) of lithium-ion batteries is essential for ensuring the safety, efficiency, and lifecycle management of electric vehicles (EVs). Although data-driven approaches have become the mainstream solution for SOH estimation, most existing studies rely heavily on laboratory datasets collected under controlled and idealized conditions. Such datasets fail to capture the stochastic characteristics of real-world vehicle operation, including fragmented charging behaviors, varying environmental conditions, and significant sensor noise. Moreover, single deep learning architectures often struggle to simultaneously model the long-term temporal evolution of battery degradation and the complex multivariate correlations among operational variables. To address these challenges, this study proposes a hybrid neural network framework termed Long Short-Term Memory (LSTM)-improved iTransformer, which integrates the temporal modeling capability of LSTM networks with the multivariate feature interaction ability of an improved inverted Transformer architecture. In addition, a high-fidelity dataset was constructed using operational data collected from ten real-world electric vehicles. To simulate a realistic cloud-deployment scenario, a strict cross-vehicle validation strategy was adopted, where data from seven vehicles were used for model training and data from three entirely unseen vehicles were reserved for testing. The experimental results demonstrate that the proposed framework significantly outperforms conventional baseline models. In the multi-vehicle experiment, the model achieved a root mean square error (RMSE) of 1.18% and a coefficient of determination (R²) of 0.97, indicating promising cross-vehicle prediction performance within the available fleet. Furthermore, in the single-vehicle robustness experiment with limited training data, the proposed model achieved the lowest prediction error with an RMSE of 0.0045% and an MAE of 0.0034%, demonstrating superior accuracy and robustness compared with baseline models. These results suggest that the proposed method is a promising solution for battery health monitoring based on real-world operational data. Full article

With the continuous growth of global population and increasing pressure on food security, the transformation toward precise and intelligent agricultural production has become an inevitable trend. In this context, accurate identification of field weeds is crucial for improving crop yields and reducing agricultural inputs. However, agricultural Internet of Things (IoT) edge devices are generally subject to strict constraints in terms of power consumption, storage, and real-time performance. Existing lightweight convolutional neural networks often struggle to simultaneously achieve high accuracy and low resource consumption for fine-grained weed identification tasks. To address this challenge, this paper proposes a hardware aware lightweight convolutional neural network named Weed-Aware Ghost Enhanced Network (WAGENet) for microcontroller deployment. The network synergistically integrates Ghost low-cost feature generation, Mobile Inverted Bottleneck Convolution (MBConv) for deep semantic extraction, Squeeze and Excitation (SE) and Coordinate Attention (CA) dual attention mechanisms for channel space joint calibration, and Atrous Spatial Pyramid Pooling (ASPP) for multi-scale context fusion. It constructs a progressive feature abstraction system from shallow textures to high-level semantics. On the public DeepWeeds dataset, WAGENet achieves 95.71% classification accuracy and 93.80% F1 score with only 0.163 M parameters and 2.43 × 10⁸ multiply accumulate operations (MACC), attaining a parameter efficiency of 587.19%/M and significantly outperforming existing mainstream lightweight models. The model has been successfully deployed on the STM32H7B3I microcontroller development board, achieving a single inference latency of 94.63 ms, an internal Flash footprint of only 686.95 KiB, and a single inference energy consumption of 41.45 mJ. Experimental results demonstrate that WAGENet achieves a trade off among accuracy, latency, and energy consumption under strict resource constraints, providing a reproducible microcontroller deployment paradigm for battery powered field robots, drones, and other agricultural IoT edge devices. Full article

In this study, parameter estimation-assisted sensorless control methods are proposed for compressor motors. As sensorless control strategies, rotating high-frequency injection (RHFI), pulsating high-frequency injection (RHFI), and an adaptive-gain sliding mode observer (AG-SMO) are employed. During startup, HFI-based methods are utilized, whereas AG-SMO is activated under steady-state operating conditions. To mitigate parameter variations and inverter nonlinearities, Adaline Neural Network (ANN), Recursive Least Squares (RLS), and Extended Kalman Filter (EKF) algorithms are integrated for the real-time estimation of stator resistance and dead-time voltage. The proposed framework is validated through both simulation and experimental studies on a 30 W, 20 V interior permanent magnet motor commonly used in compressor applications. The results demonstrate that sensorless control algorithms alone provide robust operation, while the incorporation of parameter estimation effectively eliminates stability issues and ensures reliable transitions from low to high speeds. Comparative analysis reveals that ANN has a simple structure, RLS achieves faster convergence, and EKF provides smoother estimates under noisy conditions. Overall, the integration of sensorless control algorithms with ANN/RLS/EKF-based parameter estimation and dead-time compensation offers a cost-effective and reliable solution for high-performance compressor applications. Full article

This paper investigates the distributed formation control problem for nonlinear multiagent systems subject to full-state constraints and proposes a predefined-time neural adaptive control scheme based on a nonlinear mapping technique. To handle the time-varying asymmetric constraints on system states, a smooth and invertible nonlinear mapping function is introduced to transform the original constrained states into unconstrained variables, thereby eliminating the dependence on initial conditions typically required by traditional barrier Lyapunov functions. Within this transformed framework, a predefined-time distributed formation control law is developed, which guarantees that all followers converge to the desired formation configuration and track the leader’s trajectory within a user-specified time upper bound, independent of the initial states. Radial basis function neural networks are employed to approximate the unknown nonlinear dynamics of each agent, and adaptive laws are designed to update the network weights online. Theoretical analysis shows that all closed-loop signals remain bounded, the original system states strictly stay within their prescribed constraint boundaries at all times, and the formation tracking errors converge to a small neighborhood of the origin within the predefined time. Numerical simulations validate the effectiveness of the proposed method, demonstrating faster convergence, higher steady-state accuracy, and improved robustness to initial conditions compared to existing control approaches. Full article

Although laser heterodyne interferometric sensing systems offer exceptional theoretical resolution, practical precision is constrained by intractable nonlinear errors stemming from optical imperfections. Conventional compensation methods suffer from hardware dependency, complexity, and performance degradation under low signal-to-noise ratios (SNR). To address this, we propose a precision calibration method using a self-supervised Physics-Informed Neural Network (PINN) guided by frequency-domain priors with harmonic distribution characteristics. This approach establishes a robust compensation model by inverting equivalent parameter sets and error curves in a single step. Specifically, leveraging high-precision displacement references, the method extracts measurement residuals containing periodic physical features. Subsequently, it integrates frequency-domain priors into a physically constrained network architecture: theoretical frequency characteristics construct masks to generate high-confidence pseudo-labels, while the error equation is recast as a differentiable physical layer imposing explicit hard constraints during forward propagation. This mechanism enables precise identification of the system’s nonlinear physical properties against high background noise. Experimental results show that the root-mean-square (RMS) value of the nonlinear error was reduced from 1.90 nm to 0.23 nm, with a compensation rate reaching up to 88.13%. This method provides a reliable framework for the intelligent calibration and error self-characterization of heterodyne interferometric industrial sensors in the field of precision metrology sensors. Full article

In magnetized casing environments, casing-induced magnetic interference can significantly reduce azimuth calculation accuracy and compromise the reliability of downhole tool spatial localization. To address this issue, this study proposes a magnetic-field-based spatial localization method for downhole tools in magnetized casing environments. First, a joint azimuth correction framework is developed. This framework combines ellipse fitting for radial distortion correction, single-axis multi-station analysis (MSA) for axial interference suppression, and a radial basis function neural network (RBFNN) for residual nonlinear error compensation. Subsequently, based on corrected azimuth information, the magnetic field distribution around the magnetized casing is analyzed through theoretical modeling and finite element simulation, and a cosine-type azimuthal response model is established. On this basis, a minimum-residual localization model is constructed to invert the measurement-point height and radial distance. The results show that the proposed correction framework effectively improves azimuth calculation accuracy, with the average RMSE reduced to 0.0371° after RBFNN compensation. In addition, the inverted height and radial distance show good consistency with the experimental values, demonstrating the effectiveness of the proposed localization method. This study provides an effective approach for spatial localization in complex downhole magnetic environments. Full article