Traditional network switching involves modifying the distribution network’s topology to minimize the dissipated power based on consumer demand. Multiple kinds of loads and their stochastic nature introduce variations in power dissipation owing to the uncertain reliance on the precision of the load forecast. Despite this, some existing models overlook consumer diversity and uncertainty. Even studies that account for these factors often neglect the impact of temperature fluctuations on power usage and network structure. To address this gap, this present article proposes a novel network reconfiguration approach that simultaneously considers load type, stochastic behavior, ambient temperature, linear modeling, and line limit calculations. By incorporating ambient temperature effects into stochastic load modeling, the proposed method enhances analysis accuracy. Additionally, the use of a linear model ensures computational efficiency and precise results, while explicit line limit calculations guarantee reliable network operation. The findings reveal that small systems maintain stable configurations with temperature increases up to 50%, but experience significant power losses and voltage drops when temperatures rise by 60%–100%, leading to topology adjustments. In larger networks, reconfiguration becomes ineffective for temperature increases beyond 20%, requiring line capacity upgrades. Furthermore, while some actual grids remain stable under temperature variations up to 70%, others show topology changes with only a 10% temperature increase, highlighting the substantial impact of temperature fluctuations on system performance.
To ensure the desired higher voltage conversion ratio (VCR) from fuel cell stacks, nonisolated high-gain dc-dc converters play a crucial role in electric vehicles (EVs). High-gain converters are essential because EVs often require a wide range of higher voltage levels to meet the demand for the dc bus, which drives the traction motor and other modules in a vehicle. To mitigate the use of extreme duty ratios in conventional high-gain converters, a triple-switch, switched-inductor, switched-capacitor (SC) dc-dc converter is proposed. This converter integrates an additional auxiliary switch, a simple resonant cell, and a voltage multiplier cell (VMC). These integrations provide the desired VCR while all switches operate at a nominal duty ratio, ensuring improved efficiency and minimizing switching losses by maintaining soft-switching ability. Furthermore, the proposed converter facilitates zero current switching (ZCS) turn-on operation for all switches through the utilization of an auxiliary inductor, an auxiliary capacitor, and two diodes. The operation of the proposed converter is analyzed in continuous current mode, steady-state gain, and efficiency analysis are discussed in detail. The proposed converter is compared with the existing converters and a 200 W laboratory prototype is developed to validate the theoretical analysis or to prove the effectiveness of the converter.
Correct use of home appliances is intended to avoid property damage and life unsafety. Also, serious fire accidents can affect the safety of one’s neighborhood. Therefore, this study uses the Internet of Things (IoT) platform to build an intelligent system that aims to detect the temperature rise of home appliances. It is employed to monitor the usage states of home appliances in real time. Hereby, the microcontroller, node microcontrol unit (NodeMCU-32S), is used to develop an IoT platform combined with sensors for measuring ac voltage, current, and ambient temperature so that the electrical power and temperature rise of load (e.g., electric motor) can be detected. Meanwhile, the smart measurement system is integrated with mobile devices to upload the detected datasets to the Google cloud database system. Moreover, the verification of feasibility and soundness of a system model is performed by using the Petri net tool, WoPeD, for the purpose of eliminating the improper states to optimize the system performance. Finally, the experimental results show that the proposed IoT-enabled system has a promising precision of 94.17% and a recall of 92.26%, which obviously outperforms other existing state-of-the-art systems.
In this work, a meander line (ML) periodic leaky wave antenna (LWA) is proposed with gain improvement and broadside scanning capability. A longer ML length is taken in a fixed unit cell period to improve the gain. A longer ML brings more space harmonics (SHs) in the radiation region. Therefore, the geometry of the ML is modified to obtain impedance matching at SHs transition frequencies at 8.35 and 10.45 GHz. The antenna covers the frequency range from 8 to 11 GHz in the X-band with the optimized broadside frequency (fB) at 9.4 GHz. A maximum gain of 15.5 dBi is achieved in 6.25λ0 antenna length. This antenna is a potential candidate for use in modern wireless communication where high gain is required in a small footprint area.
Stator interturn fault (ITF) is the most common failure in electrical machines; if no prompt detection is implemented, it can cause catastrophic results. This work proposes a novel method in permanent magnet synchronous machine (PMSM) drives to detect the ITF, which is insular to speed and load variations. The proposed ITF technique is based on negative-sequence instantaneous reactive power (IRP) distortions. The sensorless control of the PMSM drive, while using field-oriented technique, uses the voltage and current information for rotor position estimation. This serves the dual purpose of controlling the drive and also in developing the diagnostic technique. The IRP distortion is calculated from dq-reference frame voltage distortions, which are estimated using Luenberger observer and dq-reference frame current distortions. The novel fault indicator is calculated based on the vector magnitude of dc components obtained from negative-sequence IRP distortions, which is insular to various speed and load conditions of the drive. The proposed ITF detection technique is experimentally validated under varying load and speed conditions of the sensorless field-oriented controlled (FOC) PMSM drive. Further, a comparison of the proposed ITF detection scheme with the dq-reference frame current residuals technique shows the superiority of the proposed ITF detection scheme under various speed and load conditions of the PMSM drive scheme; furthermore, the reliability of the proposed ITF detection technique under various noise conditions is also verified.
Satellite terrestrial edge computing network (STECN) has significant potential application deploying mobile edge computing (MEC) in offloading the computational tasks by the heterogeneous Internet of Thing (IoT) terminals under terrestrial network (TN). In this article, we present a methodological approach toward modeling the simulation environment for a novel multitier multiconstellation STECN where the IoT user equipment (UE) from both aviation space and ground will offload computing tasks into the edge satellite network for processing. We designed a network model, a communication and coverage time model and a computing model under the proposed STECN. We proposed two algorithms with three offloading schemes addressing variability in latency tolerance by heterogeneous UEs. We optimized the allocation of communication and computing resources by the satellites under the model by adopting a deep deterministic policy gradient (DDPG) algorithm with an actor-critic network for training and learning. We approached simulation modeling by designing and developing relevant modules, simulation architecture, and workflow. We incorporated techniques to fine tune the training system under specific evaluation matrices so that our simulation model can be followed by others in the domain.
The CMOS-based neuromorphic computing system (NCS) face significant challenges, such as increasing energy usage and vast area footprints, surpassing the efficiency of biological brains. Spin transfer torque magnetic tunnel junction (STT-MTJ), a type of spin device@comm offers convenient benefits, including nonvolatility, increased energy efficiency, increased speed of operation, and compatibility with CMOS, making them ideal for energy-efficient spiking NCSs that exhibit neuronal behavior. However, the high energy consumption in spintronic-based NCS, primarily due to the high write current required for MTJ switching, remains a significant challenge as neurons in these systems tend to stay active longer than necessary. To address this challenge, we introduce a novel hybrid STT-MTJ/CMOS write terminate circuit (SM-WTC) that efficiently terminates the MTJ current efficiently after MTJ-state switches, significantly improving energy consumption and speed by 2.6× and 2.3×, compared to conventional NCSs. The proposed SM-WTC technique achieves energy consumption reductions of 52.7%, 58.3%, and 62.18% compared to prior work in real-time sensing (RTS) circuit, common-mode tracking and terminating circuit (CM-TTC), and conventional-NCS, respectively. A Cadence Virtuoso simulation using 65-nm CMOS technology has been used to evaluate the proposed circuit. Furthermore, SM-WTC-based NCS achieves a 67.2% improvement in energy-delay product (EDP) over conventional NCS for image edge detection. These advancements position SM-WTC as a commercially viable solution for next-generation artificial intelligence (AI) accelerators and brain-inspired computing architecture.
The classical orthogonal frequency division multiplexing (OFDM) systems gained significant new dimensions with the introduction of index modulation (IM) schemes. However, reduced data rates are the drawback in IM-based systems when implemented using higher-order modulation techniques. Hence, to improve the data rate, we proposed a new OFDM-IM system by varying the inactive subcarriers in in-phase and quadrature-phase in every subblock, namely, a dual-mode homogenous OFDM-IM (DMH OFDM-IM) system. Furthermore, we introduce a novel noise power and signal-to-noise ratio (SNR) estimation algorithm for the proposed system, which operates over a Nakagami-m fading channel. The proposed estimation algorithm makes use of nulled subcarriers available in every subblock of the proposed system to estimate noise power. The introduced estimator is both spectral and energy efficient as it uses inactive subcarriers that carry no energy. Simulation results emphasize that the developed estimator achieves lower noise power and estimates the SNR at an ideal value in contrast to the existing estimators of OFDM system. Moreover, differential noise power (DNP) is determined for the proposed system (DMH OFDM-IM) to track channel variations effectively.
In this article, a novel bidirectional boost-side interleaved switched boost (BBSISB) multiport converter is proposed. The boost-side interleaving (BSI) of the converter improves its performance in terms of input current ripple and reliability in medium-power applications. The BSI reduces the input current ripple (i.e., zero at 50% of duty ratio), and it also provides an open circuit (OC) switch fault-tolerance index (FTI) of 100% at the load end. The switched-boost action topology uses time multiplexing of boost and buck switches to produce regulated voltage at output ports, whereas the boost-side phase interleaving provides the paralleling of large input boost current with reduced current ripple. The bidirectional capability of the converter enables the power flow from the battery to the load end in the absence of an input supply. The performance analysis of the proposed converter with respect to switching loss, conduction loss, FTI, current ripple, and cost is performed and compared with similar converters. An experimental setup of the BBSISB multiport converter is developed in the laboratory with a constant-current, constant-voltage (CC-CV) mode of charging of the battery to verify the derived analytical results.
The relationship between the impedances of the paths in which power flows determines the amount of power flowing in those paths. The distributed series impedances (DSIs) are devices that can inject impedance in these paths, enabling power flow control to achieve many positive objectives. This article presents the results of a comprehensive literature review on using DSIs in smart grids. A bibliometric assessment is performed for the publications included in the review, covering the annual number of publications, the top five most cited journals, and the top five most cited publications. Moreover, the research is summarized into three main themes: developing DSI technology, using DSIs to solve various power system problems, and optimizing DSI installations. After this, research trends addressing how the research foci have evolved, and areas, where no or little research has been done, are discussed. Finally, the proposed areas for future research are presented.
Inverter-based resources (IBRs) play a crucial role in microgrid operation due to their ability to provide power conversion and control functions. Ensuring compliance with standard and grid codes, however, poses a significant challenge in IBR-dominated microgrids. This challenge is exacerbated by the lack of a comprehensive set of instructions tailored specifically for IBR-dominated microgrids, which ideally should consolidate all requirements into a single, centralized document. This article focuses on Canadian grid codes, specifically those in Ontario, due to the growing integration of renewable energy sources and the regulatory framework guiding microgrid development in the region. By analyzing Ontario grid codes, this study aims to provide insights relevant to IBR-dominated microgrid projects within similar regulatory contexts worldwide. A comprehensive analysis of the pertinent standards and grid code provisions is conducted, highlighting the required and recommended protection schemes. Additionally, a case study is presented to demonstrate the practical implementation of protection plan design in compliance with Ontario grid codes. Through this study, insights are provided into the complex interplay between standard and grid code compliance, as well as protection plan design considerations in IBR-dominated microgrids.
Altering the flow of power along branch reconfiguration of radial distribution feeders and mitigating the reactive power component through optimal shunt capacitor placement are proven methods for reducing energy losses in distribution systems. However, it is crucial to recognize that variations in load demand can significantly impact the magnitude of these energy losses and reactive power installation costs, potentially influencing the optimal placement of capacitors and the strategy for branch switching. Therefore, accounting for fluctuations in power demand when reconfiguring the network and positioning capacitors is of paramount importance. Nevertheless, incorporating changes in power demand while simultaneously optimizing branch configurations and addressing reactive power in radial feeders can complicate the computational aspects of the problem, leading to increased processing times. Conversely, disregarding the consumption patterns on the demand side can result in inaccurate calculations of distribution losses and related costs. Consequently, this study delves into the influence of demand patterns on the problem of network topology modification and capacitor assignment considering capacitor and switches investment. It aims to determine whether taking into account load variability is merely an option or an indispensable factor in minimizing the cost of energy losses, switching expenses, and reactive power installation budget via the placement of capacitors and altering the topology of the network. The analysis was carried out on multiple distribution grids using a classical optimization means known as a mathematical programming language (AMPL).
The design of a sparse sampler for effective coarray-domain processing of underdetermined direction-of-arrival (DOA) estimation is of prime research interest. Toward this, several sparse samplers have been proposed, among which co-prime samplers have garnered significant attention due to their closed-form expression for sampler element positioning and minimal susceptibility to mutual coupling effects. In this article, an enhanced co-prime sampler (ECPS) is proposed by concatenating the two operations on subsamplers, involving the scaling of one subsampler and shifting of another subsampler within the hold of the co-prime property. The proposed ECPS offers several advantages over existing samplers, including increased consecutive lags, reduced holes in the difference coarray, larger physical aperture, enhanced sparsity, and reduced susceptibility to mutual coupling effects. These improvements lead to high consecutive degrees-of-freedom (cDOF) in detecting more targets than the number of physical elements in the sampler with high estimation accuracy. The superiority and efficacy of the proposed ECPS are analyzed and validated with several simulation studies.
High-definition (HD) maps play an important role in autonomous driving by providing vehicles with localization functionality, environmental information, and support its mission and motion planning. In an HD map, road network extraction/annotation is a crucial feature that helps the autonomous vehicle navigate and keeps it within the safe driving zone. While road network extraction traditionally focuses on motorways and their boundaries, the emergence of small-scale autonomous vehicles, such as delivery and service robots, has shifted attention to sidewalks. Sidewalks are critical for safe and efficient navigation in residential and urban areas, yet automated methods for sidewalk extraction remain underexplored. To address this gap, this article proposes a sidewalk extraction method on aerial images using deep learning with the transfer learning technique. A path-planning algorithm-based refinement method is also proposed to further refine the extracted sidewalk. The proposed method can precisely extract sidewalks from aerial photographs and fix sidewalk discontinuity issues caused by occlusions. A sidewalk dataset is also explicitly designed for sidewalk extraction and semantic segmentation research. This article’s work is meant to fill the sidewalk extraction gap for road network extraction.
Weather conditions directly affect sectors such as agriculture and transport. With climate change, unpredictability is increasing and traditional calculation methods may not be sufficient. In addition to some statistical methods, machine learning algorithms are also used for weather forecasting. This study attempts to classify precipitation using machine learning algorithms on selected meteorological data. The models used are K-nearest neighbors (KNNs), support vector machine (SVM), and multilayer perceptron (MLP). These models were implemented on four different open-source and free data mining platforms. These platforms are Altair AI Studio (formerly Rapidminer), Knime, Orange, and Weka. The dataset includes parameters such as pressure, temperature, humidity, number of rainy days, cloudiness rate, and year and month information. According to the values of these parameters, the data were classified as less rainy, rainy, and very rainy.
High impedance faults (HIFs) present significant challenges in power systems, particularly when an electrical wire contacts a high-resistance material, leading to low currents that are difficult for traditional relays to detect. With the increasing integration of photovoltaic (PV) systems, these challenges are exacerbated due to the complex behavior of PV-generated signals. This study aims to enhance the detection of HIFs in PV-integrated systems using advanced machine learning techniques. The approach employs various classifiers, including artificial neural networks, support vector machines (SVMs), decision trees, and random forest (RF) to improve fault identification accuracy. A MATLAB/SIMULINK simulation was conducted on an IEEE 13-bus system with a 300-kW solar PV plant. The discrete wavelet transform (DWT) with the db4 wavelet was used for feature extraction, focusing on phase energy values. The classifiers were evaluated under different scenarios, such as normal operation, load switching (LS), capacitor switching (CS), HIF, and line-to-ground (LG) faults. The RF classifier outperformed others, achieving a fault detection accuracy of 99.4083%, demonstrating its robustness in adapting to various fault conditions. The Naive Bayes (NB), multilayer perceptron (MLP), and logistic regression (LR) classifiers achieved lower accuracies of 78.6982%, 76.9231%, and 80.4734%, respectively. These results indicate a significant improvement in fault detection capability, enhancing the stability, reliability, and resilience of electrical grids integrated with PV systems. The findings suggest that the RF classifier is highly effective for HIF detection, which is crucial for the protection and efficient operation of modern power grids with high renewable energy penetration.
In this article, a handwriting teaching system based on artificial intelligence (AI) edge computing technology is proposed. The proposed system combines gesture tracking, gesture recognition, and other related AI technologies. Additionally, the development platform for AI edge computing in this system is developed to teach handwriting and practice drawing. The proposed system is composed of a teacher-end teaching host and several student-end AI edge computing smart devices. The student-end AI edge computing smart device incorporates virtual drawing and writing, finger digital computing teaching, a virtual keyboard, virtual sliding, and sleep prevention warnings. The teacher-end teaching host allows the teacher to conduct a teaching course. Moreover, the teacher-end teaching host and the student smart device can simultaneously display images on a large screen to facilitate teaching demonstrations. Furthermore, this system has a comprehensive data storage cloud platform, which can record the data uploaded by each student to a storage cloud platform to facilitate teaching evaluations. This work differs from traditional handwriting and painting technique studies in the classroom, and the AI virtual drawing technology proposed in this work can produce impressive visual effects for visual media, including animation, graphics, and text.
In this work, we propose a joint force control framework for the stable motion of a humanoid biped robot. In motion planning, nonlinear centroid dynamics is used to generate gait on uneven terrain, which overcomes the limitation of a linear inverted pendulum (IP) model on centroid height. The motion control layer combines multipriority inverse kinematics (MPIK) and multipriority dynamic control (MPDC). The MPIK uses a multipriority inverse kinematics numerical iteration algorithm to calculate joint position command. The MPDC uses a multipriority iterative optimization method based on the task-space dynamics model on the forward path, which does not need preallocation or preoptimization of contact force, does not explicitly control the movement of center of mass (CoM), and tries its best to complete high-priority tasks. Finally, a stable joint compliance force control framework is built, and the introduction of kinematic error information ensures the accurate position tracking of the force control system. The results show that the control strategy completes the task of climbing stairs well and shows a certain antidisturbance ability in standing still and variable speed walking. The maximum disturbance in the sagittal plane can reach 50 N⋅s (achieved solely by adjusting the position of the pressure center and without using the step stability strategy).
The objective of this work is to compare the performances of online and offline wide-area control system designs by considering the limited number of generators. The design of the controller feedback gain matrix in both techniques is achieved by the state feedback control technique. Both designs are implemented with a limited number of generators. However, the required structure of the feedback gain matrix in offline mode can be accomplished by using the structurally constrained H2-norm optimization. On the other hand, the required gain matrix in online mode can be designed with the help of a real-time control input matrix, right and left eigenvectors. The phasor measurement units (PMUs) data is used in both designs. Both the state vector and the feedback gain matrix are computed in real-time in online mode. Whereas in offline mode, only the state vector is obtained from PMU measurements and the feedback gain matrix can be designed with the help of available offline data of a particular test system. The merits and demerits of both designs are explained in detail by considering different aspects. The comparison of the performances of both designs is illustrated in MATLAB/Simulink environment by considering the IEEE-68 bus test system.
