IEEJ Transactions on Electrical and Electronic Engineering最新文献

Highly sensitive magnetic field sensors are essential for applications such as biomedical diagnostics and electromagnetic compatibility (EMC) assessments. This study focuses on optimizing a GHz-range thin-film magnetic sensor by incorporating a slit pattern in the magnetic layer, which improves impedance matching and current distribution. Amorphous CoNbZr thin-film sensors with slit widths ranging from 0 μm (no slit) to 50 μm were fabricated. Among these, the 10 μm slit design exhibited the best impedance matching and magnetic field sensitivity. Additionally, a carrier suppression circuit was implemented to reduce phase noise, resulting in improved sensitivity, achieving a minimum detectable field of 1.41 × 10⁻¹¹ T/√Hz. The sensor was employed for precise near-field measurements in wireless power transfer (WPT) systems, where it outperformed conventional loop antennas. Experimental results demonstrated superior sensitivity and enhanced spatial resolution, particularly near the ferromagnetic resonance under specific DC bias and carrier frequency conditions. These findings highlight the potential of the optimized sensor for improving electromagnetic interference (EMI) and EMC assessments. © 2025 The Author(s). IEEJ Transactions on Electrical and Electronic Engineering published by Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

The quantity and quality of rice output are significantly impacted by rice diseases. The accurate and fast diagnosis of rice diseases is the key to avoiding these negative effects. Recent remarkable advancements in the fields of the Internet of Things (IoT) and artificial intelligence (AI) have given rise to a novel concept known as Artificial Intelligence of Things (AIoT). In this paper, an AIoT-based system has been designed for the efficient classification of rice leaf diseases. Machine learning models have been built based on deep convolutional neural networks such as YOLO and ResNet and evaluate the capability of distinguishing healthy cases from three types of rice leaf diseases, including leaf blast, hispa, and brown spot, and healthy. Each machine learning model provides quite good classification results; however, there are still some confusion cases between types of diseases. The most optimal machine learning model was used to deploy on IoT platforms, including servers and web applications. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

Diagnostic methods based on electrical network theory are commonly employed for locating corrosion in grounding grids. Due to the relatively small number of nodes available for measurement compared to the total number of nodes, the resulting diagnostic equations for corrosion are underdetermined and nonlinear, making it challenging to accurately determine the location and extent of corrosion in branches. To address this issue, this study proposes an improved Wild Horse Optimization algorithm (MCP-IWHO) that incorporates MCP regularization for solving corrosion equations. This enhanced algorithm achieves a balance between global and local search capabilities and controls the sparsity of the solution to reduce deviations. Simulation calculations were conducted using the actual grounding grid of a 110 kV substation in a specific region as a case study. The results demonstrate that this algorithm can accurately determine the position and severity of corrosion in branches, outperforming both the standard Wild Horse Optimization algorithm (WHO) and traditional optimization algorithms in terms of error minimization and convergence speed. This provides a more efficient and reliable method for diagnosing corrosion in underground grounding grids. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

For the wind power generation system, the fault ride-through control strategy of the inverter makes the fault current exhibit nonlinearity and non-stationarity, which poses challenges to the traditional pilot protection that relies on the accurate measurement of amplitude or phase. This paper proposes a pilot protection based on mutual information (MI). Firstly, the current measured at both sides of the transmission line is reconstructed to mitigate the impact of phase difference. After that, the normalized MI value is defined as the index for distinguishing between internal and external faults. Simulation results show that the proposed method accurately discriminates faults under various conditions. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

As an intuitive form of interaction, gestures are widely used in fields like smart live streaming and smart homes, significantly improving user efficiency and the convenience of human-computer interaction. However, existing gesture recognition methods often suffer from large model sizes and high computational complexity, making them unsuitable for real-time deployment. To address these issues, we propose GF-YOLOv8, a lightweight gesture recognition model based on YOLOv8 (where ‘GF’ stands for the initials of two optimization modules: G represents the GB-C2f module, and F represents the FFPN network), designed to reduce parameters while maintaining detection accuracy. First, by designing the GB-C2f module, which integrates the lightweight network GhostNet, the model significantly reduces computation and parameters with minimal loss in accuracy. Second, a fusion feature pyramid network (FFPN), based on ASFF and AFPN algorithms, and enhances the interaction between non-adjacent layers to improve gesture feature perception. Compared to the original YOLOv8 model, the proposed method demonstrates significant improvements on both public and self-constructed datasets: GFLOPs and parameter count are reduced by 46.9% and 44.9%, respectively, with the mAP value remaining almost unchanged. The improved model enables efficient and accurate gesture recognition on mobile and embedded devices, achieving a substantial reduction in model size without compromising accuracy. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

The original simulation verified the fuel-saving effect of the energy storage system for adjusting the load factor of generators to optimal conditions. Through the analysis of the simulation results, it was revealed that the energy storage system showed the potential to save the fuel consumption of generators by more than 600 l in 7 days under certain simulated conditions. This saving was mainly attributed to the reduction of the parallel running time of the generators due to the discharging of the energy storage system. It was also found that the fuel saving effect could be maximized by designing the rated power of the converter appropriately for the energy storage system. © 2025 The Author(s). IEEJ Transactions on Electrical and Electronic Engineering published by Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

Aiming at the problems of small reluctance torque and poor weak magnetic capability of SPM motor and excessive torque ripple of IPM motor, this paper designs a new type of permanent magnet synchronous motor with an asymmetric hybrid structure according to the magnetic field shift effect. Firstly, establishing the finite element model of this motor to explore the effect of permanent magnet and magnetic barrier parameters on the magnetic field shift effect and torque ripple. Subsequently, traditional IPM motors and AHRPM motors were optimized and compared under the same objective function and conditions. The results showed that AHRPM motors can output greater electromagnetic torque and lower torque ripple with fewer permanent magnets, while also offering better speed control performance and efficiency than traditional IPM motors. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

In normal operation, the ultra-high voltage transmission line induces current and power loss on the overhead ground line. To demonstrate the law of induced current and loss from the ground line of a 750 kV double overhead wire in the same tower, ATP-EMTP electromagnetic transient simulation software was used to calculate the model. The numerical study of induced current and loss using optical fiber composite overhead ground wire (OPGW) and conventional ground wire grounded tower by tower. Five factors were evaluated for their effects on ground wire current and loss. The results show that OPGW and ordinary ground wire induced current and loss increase with operating voltage and line capacity, but decrease with line length, conductor-ground line distance, and horizontal spacing between lines. Transmission capacity, operating voltage, line length, conductor-ground wire distance, and horizontal distance between two circuits were examined using multiple fitting to see how they affect overhead ground wire induced current and loss. To improve line architecture and reduce ground wire loss, quantifiable mathematical formulae for induced current and ground wire loss were created. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

This paper presents a data-driven design methodology for a gain-scheduled controller applied to a controlled system modeled as a linear parameter-varying (LPV) system. In the proposed approach, a controller is directly synthesized from time series data on the inputs and states of the LPV system. Through the methodology, a controller that ensures closed-loop stability is derived by solving a convex optimization problem. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

During the current breaking period of low-voltage DC circuit breakers, their contacts and arc extinction system are eroded by arc combustion. The morphology of the arc plate significantly changes after arc erosion. On this basis, a service times prediction model of low-voltage DC circuit breaker is proposed. First, by observing the morphological changes of the arc plates, it is found that a bright metal fusion trace appears on their surface. With increasing occurrence of arc erosion, significant material loss appears at the top of the arc plates. Then, the grayscale transformation enhancement and region growth algorithms are proposed to preprocess and segment the morphological image of the arc plate to obtain the metal fusion trace. An image mask processing method is proposed to extract the contour of the arc plate to obtain the material loss area. Finally, the two morphological features of the arc plate are used as degradation indicators for low-voltage DC circuit breakers, establishing a service times prediction model based on Gaussian process regression. Through experimental verification, the relative errors between the predicted and true values of the current service times are verified to be less than 20%. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.