For microgrid in islanded operation, due to the effects of mismatched line impedance, reactive power could not be shared accurately with conventional droop method. To improve sharing accuracy, this paper proposes an improved control The proposed method mainly includes two important operations: error reduction operation and voltage recovery operation. accuracy is by which activated low-bandwidth synchronization signals. However, will result a decrease output amplitude. Therefore, compensate...

The COVID-19 coronavirus has spread rapidly around the world and caused global panic. Chest CT images play a major role in confirming positive patients. computer aided diagnosis of from based on artificial intelligence have been developed deployed some hospitals. But environmental influences movement lung will affect image quality, causing parenchyma pneumonia area unclear images. Therefore, performance COVID-19's diagnostic algorithm is reduced. If chest are reconstructed, accuracy may be...

Vehicle detection is a challenging problem in autonomous driving systems, due to its large structural and appearance variations. In this paper, we propose novel vehicle scheme based on multi-task deep convolutional neural networks (CNNs) region-of-interest (RoI) voting. the design of CNN architecture, enrich supervised information with subcategory, region overlap, bounding-box regression, category each training RoI as learning framework. This allows model share visual knowledge among...

Large language models (LLMs) offer significant potential as tools to support an expanding range of decision-making tasks. However, given their training on human (created) data, LLMs can inherit both societal biases against protected groups, well be subject cognitive bias. Such human-like bias impede fair and explainable decisions made with LLM assistance. Our work introduces BiasBuster, a framework designed uncover, evaluate, mitigate in LLMs, particularly high-stakes Inspired by prior...

Summary A three‐dimensional (3D) reconstruction method of structured light based on photoelastic fringes was proposed in this research. The are produced by both simulation and a polycarbonate disk under diametric compression load. Six projected onto an object using the six‐step phase shifting technique. Therefore, isochromatic image is calculated. After unwrapping, can be used for 3D reconstruction. In order to verify effectiveness method, two experiment devices were built projector...

Three-dimensional (3-D) gravity inversion obtains the density distribution of subsurface geological bodies through observed anomalies. Recently, data-driven methods based on deep neural networks (DNNs) have received considerable attention for geophysical inverse problems. They proved to be superior physics-driven that suffer from nonuniqueness issues and high computational costs. Several deep-learning strategies been developed modeling are mainly applicable two-dimensional (2-D) imaging....

Tremendous efforts have been put forth on predicting pedestrian trajectory with generative models to accommodate uncertainty and multi-modality in human behaviors. An individual's inherent uncertainty, e.g., change of destination, can be masked by complex patterns resulting from the movements interacting pedestrians. However, latent variable-based often entangle such complexity, leading limited either expressivity or predictive diversity. In this work, we propose separately model these two...

ABSTRACT Current electric vehicle trajectory prediction fails to fully consider the interaction between target and other vehicles, resulting in poor results. In order solve this problem, paper proposes a hybrid attention transformer network (HATN), which is designed for more accurate prediction. Firstly, based on network, introduces self‐attention mechanism cross mechanism, feature embedding position encoding module as well an interactive extraction module, so achieve modelling of state...

With the development of unmanned aerial vehicle (UAV) remote sensing technology, target detection based on UAV images has increasingly become a hot spot for research. Aiming at problems many small instances, complex backgrounds and difficult feature extraction in images, we propose image algorithm called BLUR-YOLO. First, h-swish activation function is used backbone network neck to increase expressiveness model. Second, an attention mechanism (CoordAttention) added bottleneck layer network,...

Abstract Pavement segments are functionally interdependent under traffic equilibrium, leading to maintenance and rehabilitation (M&R) decisions for different segments, but it has not received significant attention in the pavement management community yet. This study developed a optimization model networks based on simultaneous network (SNO) framework multi‐agent reinforcement learning algorithm. The established was demonstrated highway real‐world, compared previously built two‐stage...

Internet streaming applications are becoming increasingly popular on mobile devices. However, receiving services devices is often constrained by their limited battery power supply. Various techniques have been proposed to save consumption devices, mainly focusing how much data transmit and transmit.

Smart phones nowadays are equipped with global positioning systems (GPS) chips to enable navigation and location-based services. A malicious app the access GPS data can easily track person who carries smart phone. People may disable module turn it on only when necessary protect their location privacy. However, in this paper, we demonstrate that an attacker is still able a by using embedded magnetometer sensor victim's phone, even disabled all time. Moreover, attack neither requests user...

To improve the operating performance of large-capacity battery pack electric vehicles during continuous charging and discharging to avoid its thermal runaway, in this paper we propose a new hybrid management system that couples PCM with liquid cooling plate microchannels. The flow direction microchannel structure bottom is designed according characteristics large axial conductivity battery, whole under charge/discharge cycles numerically simulated. results show PCM/liquid can maintain good...

In the vast sea of information, powerful ability machine learning(ML) algorithms to identify specific objects through goal guidance makes them stand out from many methods people use. As a result, ML are widely used in business decision making. This paper presents research analysis financial risk network assessment model(FRNAM)s based on artificial intelligence(AI) and algorithms. The main sources supply chain(SC) finance integrated process briefly discussed, transmission effect SC is...

Trajectory prediction is fundamental to various intelligent technologies, such as autonomous driving and robotics. The motion of pedestrians vehicles helps emergency braking, reduces collisions, improves traffic safety. Current trajectory research faces problems complex social interactions, high dynamics multi-modality. Especially, it still has limitations in long-time prediction. We propose Attention-aware Social Graph Transformer Networks for multi-modal combine Convolutional by generating...

Wireless security cameras are integral components of systems used by military installations, corporations, and, due to their increased affordability, many private homes. These commonly employ motion sensors identify that something is occurring in fields vision before starting record and notifying the property owner activity. In this paper, we discover sensing action can disclose location camera through a novel wireless localization technique call MotionCompass. short, user who aims avoid...