Journal of Electrical Engineering & Technology最新文献

When a fault in secondary feeder is occurred in MVDC distribution system, the operation method of protection devices is being required to achieve better protection coordination in order to minimize interruption region in primary feeder, considering the operation time interval of the protection devices in primary and secondary feeder before the shut-down of the main converter. Therefore, this paper presents an intelligent fault isolation device(IFID) to instantaneously clamp and consume the fault current in order to prevent shut-down of main converter in radial-type MVDC distribution system. And also, the proposed device is installed in both main converter and section converter, its operation mechanism is categorized as 4 modes such as initial operation mode, auxiliary operation mode, main operation mode and restoration mode. From the simulation results based on modeling of the IFID using PSCAD/EMTDC, it is confirmed that the proposed device is a useful tool to isolate fault section by preventing shut-down of main converter and securing operation time interval for protection coordination in MVDC and LVDC feeders. And also, it is founded that the main converter can be continuously operated and then unfaulted sections in primary feeder can be supplied as the faulted section is isolated by section converter.

In Simultaneous Localization and Mapping (SLAM) techniques, the precise estimation of the initial pose of a mobile robot presents a significant challenge. The initial pose is crucial as it can significantly reduce the accumulated errors in SLAM. Despite various advancements in pose estimation, accurately determining the initial pose in different application scenarios continues to be a complex task, often requiring techniques such as loop closure. However, loop closure detection may not always be feasible, as it is time-consuming and relies on the robot revisiting previous paths or its initial starting position. In essence, most localization methods such as SLAM critically require the initial pose to estimate the mobile robot’s pose and the more precise initial pose makes the localization process to converge quickly. Addressing these persistent challenges, this paper proposes a novel method that utilizes both image and point cloud data, allowing for easy adaptation across diverse and dynamic environments. The method integrates well-known technologies such as NetVLAD, RootSIFT, 5-Point, and Iterative Closest Point (ICP) algorithms. This approach not only addresses the initial pose estimation problem but also provides an alternative to existing landmarks, enhancing adaptability to diverse and dynamic environments. NetVLAD is utilized to find the most similar image data in stored images by comparing the image captured by the mobile robot with the stored images with pose data. The relative pose is estimated by applying the RootSIFT and 5-Point algorithm, and ICP algorithm to the found image and point cloud data, respectively. This method determines the final pose of the mobile robot by combining each relative pose extracted from image data and point cloud data through weighted integration. The effectiveness of the proposed method is verified by comparing it with existing deep learning-based pose estimation methods. This method can accurately estimate poses, including the initial pose, using much less data than existing deep learning methods, even in diverse and dynamic environments. Furthermore, this method is applicable not only when using image data alone but also when both image and point cloud data are available.

Distance relays are crucial components of power transmission systems, providing protection against faults and ensuring the stability and reliability of the electrical grid. However, under certain conditions, they may experience maloperations, leading to erroneous tripping or failure to trip during fault events. The proposed scheme incorporates a modified Mho characteristics for monitoring load impedance to improve loadability during load encroachment conditions and prevent false tripping. In addition, the pilot scheme Permissive Under reach Transfer Trip (PUTT) provides communication capabilities enhancing the relay’s accuracy in fault detection. Extensive simulations and tests are conducted using MATLAB/Simulink to validate the effectiveness of the pilot scheme in EMTP (Electro Magnetic Transient Program) environment. The results indicate that the proposed scheme achieves faster detection of remote circuit breaker operations by 520ms bypassing the Zone 2 set delay of 500ms also. The scheme offers a practical and cost-effective solution to mitigate the challenges associated with load encroachment and enhance the overall stability and reliability of power systems.

Lung nodule detection is clinically crucial but challenging and time-consuming. The development of automated segmentation approaches could be helpful. To assess the capability of deep learning methods for lung diagnosis, this paper compares recent deep learning models and evaluates their performance. We implemented several preprocessing steps, including windowing, thresholding, and resizing, to improve the image quality, adjust the image dimension suitable for the network, and focus on specific areas of interest within an image. The evaluation was conducted on the Lung Image Database Consortium (LIDC) dataset using Dice similarity coefficient (DSC) and Hausdorff distance (HD) metrics with model complexity parameters for a multifaceted comparison of the models. The experiments showed that the highest accuracy among the five chosen models (97.80% DSC and 1.29 HD) was reached by the Connected-UNets model, which also has the highest computational complexity. In this paper, we quantitatively evaluated and compared 5 deep learning models namely Salient Attention UNet, Connected-UNets, DDANet, UTNet, and EdgeNeXt. The evidence-based overview of current deep learning achievements for the clinical community investigated in this study can be useful to the research community in developing a new model and, thus, designing computer-aided detection and diagnosis (CAD) systems.

Hydrogen mobility in Korea plans to supply 850,000 units by 2030. It is necessary to expand hydrogen production and charging facilities in urban areas with good accessibility. In order to build a hydrogen charging station in an urban area, the facility site of test bed should be optimized and minimized. In this paper, a method to utilize the surplus power of urban railways is proposed such as a integrated power flow from existing railway station and the requirement of hydrogen charging systems. It includes production specifications for power converters for linking power supply systems with hydrogen charging systems such as optimal converter quantity. This paper contributes to expand hydrogen production less costly due to railway surplus power and a limited downtown space.

In this study, a Matlab Graphical User Interface (GUI) was developed to facilitate the input of design criteria in the Ansys Maxwell RMxprt module for preliminary Induction Motors (IM) design. Scanned images of the stator core plate and the slot section were used in the interface to define the slot of the IM. Image processing methods were applied to extract the geometric dimensions of the stator sheet plate from these images. The interface program successfully obtained the stator’s outer diameter, inner diameter, and the number of stator slots with a very low error rate. Once the geometric measurements were obtained, other motor design criteria were inputted, and the data was automatically transferred to the RMxprt module. RMxprt module then utilized this data for motor analysis. The analysis results were exported to Matlab for visualization. This approach aims to reduce user-induced measurement errors during the extraction of motor stator geometric measurements and provide practical data input. Furthermore, there is a need for the creation of interdisciplinary learning environments through simulation programs. In this context, to enhance the learning experiences of researchers undergoing IM training, an alternative simulation program that uses new technologies is introduced, creating enriching training environments.

This paper addresses de-identification problem of vehicle license plates in image-based datasets crucial for autonomous driving and robotics research. Traditional methods such as blurring and masking used for privacy protection reduce data quality by obscuring key features like license plates, limiting the utility of datasets. To overcome this, this paper introduces a new method that retains original data quality while anonymizing license plates by combining detection, vertices identification, and generation of virtual license plates resembling textures of the original. The approach utilizes advanced models such as YOLOv8x and Swin Transformer, enhanced by the Weighted Box Fusion ensemble technique, to improve the detection accuracy of license plates in complex indoor parking lot environments characterized by poor lighting and varied vehicle positions. This method enhances the accuracy of detecting multiple license plates in a single image, isolating images with only one vehicle for further processing. The RetinaFace model identifies the precise positions (or 4 vertices) of license plates and the perspective transformation technique replaces the original plates with virtual ones using the identified vertices. These virtual plates are randomly generated but initially do not match the original plates' texture, making them obviously artificial. To address this, a finely adjusted CycleGAN model adapts the texture of the virtual plates to closely resemble the originals. Finally, the modified plates are merged back with the original vehicle images, ensuring that the dataset retains the visual similarity to the original images while effectively making de-identification of the license plates for privacy. The effectiveness of the proposed de-identification method is validated by comparing its performance with various deep learning models in an indoor parking environment.

High voltage line-starting permanent magnet synchronous motor (HVLSPMSM) have the advantages of induction motors in terms of ease of operation and maintenance. However, due to the slot structure of the stator and rotor, the electromagnetic force composition is complex, which causes the vibration characteristics of HVLSPMSM to change. In order to analyze the influence of stator–rotor slot combinations on the vibration of HVLSPMSM. In this paper, a 10 kV HVLSPMSM is taken as an example to study. Firstly, the mathematical model of HVLSPMSM is established and the air gap magnetic flux density distribution characteristics are obtained for different stator–rotor slot combinations. The expression of the air gap electromagnetic force density (AEFD) of HVLSPMSM is derived by Maxwell tensor method, and the effect of different stator–rotor slot combinations on AEFD is revealed. Further, based on the theory of stator tooth modulation, the sources of equivalent concentrated forces (ECFs) under for different stator–rotor slot combinations are analyzed and the effect of stator–rotor slot combinations on electromagnetic vibration is determined. Next, the electromagnetic–mechanical coupled model of HVLSPMSM is established, and the electromagnetic and vibration characteristics under different stator–rotor slot combinations are solved, and the effect of stator–rotor slot combinations on motor vibration is obtained. Finally, the electromagnetic experiment is conducted to validate the analysis.

In this paper, a novel cascaded predictive control strategy, endowed with the merits of the deadbeat predictive control solution and Kalman filter framework, is proposed for unipolar sinusoidal excited switched reluctance motors. The primary aim is to augment the dynamic performance of the system while ensuring robustness across various conditions. To achieve this objective, a Kalman filter is meticulously implemented and integrated into the deadbeat predictive current control strategy (DPCC). This integration facilitates a comprehensive consideration of the system’s nonlinear dynamics and measurement noise, thereby reducing the DPCC’s reliance on precise model parameter information. This advancement significantly bolsters the robustness and flexibility of the control system in various operational contexts. Simultaneously, to enhance the dynamic performance of the speed control mechanism and robustness, a deadbeat predictive speed control strategy (DPSC) integrating a novel adaptive fading Kalman filter (AFKF) is proposed. Within the AFKF, a modified fading factor selection method is employed, which contributes to the enhanced convergence speed, thereby optimizing the dynamic performance of the DPSC strategy. Finally, detailed comparative simulations and experiments are undertaken to verify the theoretical framework’s feasibility and to substantiate the performance enhancements ascribed to the proposed control strategy.

Light electric vehicles (LEV) have become an integral part of our daily lives. In-wheel Brushless Direct Current Motor (BLDCM) is mostly preferred in the propulsion systems of LEVs. Ferromagnetic materials used in electric motors affect motor output performance as much as geometric design parameters. The main objective of this study is to investigate the effects of vibration on the driving performance of LEVs and its disruptive effects on BLDCM. Vibration effect reduces the operating life of electric motors and leads to their malfunction. In this study, the vibration effects of different stator and rotor ferromagnetic materials in-wheel BLDCM of LEVs were analyzed. It has been revealed that permanent magnets and mild steel ferromagnetic materials are factors affecting the operating performance of BLDCM. Total deformation harmonic response analyses were performed under ANSYS Workbench software to reveal the vibration effects on the specified in-wheel BLDCM. The lowest vibration effect is achieved when SmCo₅(R18) magnet is used in the rotor and M22-26G mild steel ferromagnetic material is used in the stator, where the maximum vibration of 0.086023 μm in rotor and 18.386 μm in stator are achieved. It was concluded that the most compatible materials result in the lowest vibration values in BLDCM.