Road segmentation from high-resolution visible remote sensing images provides an effective way for automatic road network forming. Recently, deep learning methods based on convolutional neural networks (CNNs) are widely applied in segmentation. However, it is a challenge most CNN-based to achieve high accuracy when processing with rich details. To handle this problem, we propose method Y-shaped (indicated as Y-Net). Y-Net contains two-arm feature extraction module and fusion module. The...

Despite recent advances in Video Object Segmentation (VOS), state-of-the-art models still face challenges with occlusion, fast motion, and long-term tracking. These difficulties often result a noticeable degradation of segmentation accuracy as the video progresses, leading to poor performance over extended periods. To tackle these persistent issues, we propose an innovative approach that enhances by introducing predictive heads. Building upon Cutie model, our method enables model predict...

The Internet of Things (IoT) has been considered as the third revolution in digital technology after computer and Internet, it brings significant advantages to smart grid. Disaster prediction prevention power-lines is one most difficult problems for electricity transmission companies. Advanced sensing communication technologies IoT can effectively avoid or reduce damage natural disasters power-lines, hence improve reliability stability power transmission. In this paper, a three-layer...

Airport traffic flow prediction is a fundamental research topic in the field of air management. Most existing works focus on single airport with temporal dynamics but fail to consider influence topological network. In this paper, novel deep learning-based framework, called network (ATFPNet), proposed capture spatial-temporal dependencies historical (departure and arrival) for multiple-step situational (network-level) arrival prediction. Firstly, considering nature distribution context...

The Internet of Things (IoT) technology plays an important role in the construction smart grid. With development grid, IoT has been deepening and extending its applications power system. In this background, paper first introduces latest progress architecture IoT, then proposes a status monitoring early warning system for distribution network based on technology. key technologies functions as well application are also described paper. can be used to display panoramic view all equipment...

With the rapid development and widespread application of Wireless Body Area Networks (WBANs), traditional centralized system architecture cannot handle massive data generated by edge devices. Meanwhile, in order to ensure security physiological privacy identity patients, this paper presents a protection strategy for Mobile Edge Computing(MEC) enhanced WBANs, which leverages blockchain-based decentralized MEC paradigm support efficient transmission information with low latency, high...

The applications of Internet Things (IoT) technology turn up in smart grid power generation, transmission, transportation, distribution and consumption gradually, it is beneficial for effective management the infrastructure. This paper introduced a transmission tower protection system based on IoT technology, which can realize classification, location early warning threats to high-voltage backbone tower, achieve full-range facilities.

A numerical method to obtain a complete solution of the dynamic response variable swept wing aircraft in course changing sweep angle is presented. Both aerodynamic and trajectory computations are included. for calculating unsteady characteristics wing/tail combinations at subsonic speeds under varying backs developed. The uses potential flow model consisting discrete vortex rings distributed over tail surfaces. set equations flight dynamics main body axes formed with deformable system...

With the development of smart grids, lots technologies are applied to solve problems power systems. The substation one-line diagram automatic generation is core Currently, traditional methods not suitable for solving problems. In this paper, we introduce newly deep learning generate automatically. Firstly, our algorithm adopts YOLOV3 method detect symbols, which a multiscale object detection algorithm, it very effective small objects like Disconnector. Secondly, PPOCR used position and...

Object tracing, especially continuous object is an important application in wireless sensor networks (WSNs). Due to the loose infrastructure, unfixed topology and limited resource, it a big challenge trace objects WSN by sensors' cooperation. In this paper, we propose energy-efficient scheme detect based on ring-architecture. Different from previous works, our solution optimizes report range computing location of phenomena each boundary face but not nodes, which reduces energy consumption...

This paper focuses on the design of a robust nonlinear autopilot for fixed-wing unmanned aerial vehicle. A disturbance observer based backstepping control method is proposed to deal with nonlinearities and uncertainties simultaneously. The baseline controllers airspeed, altitude, roll angle command tracking sideslip regulation are constructed recursively backstepping. Nonlinear observers designed obtain estimations lumped disturbances from wind aerodynamic uncertainties. As result,...

The dynamic performance of modern machine tools is more and highly emphasized. research purpose this paper to gantry machining center through modal experiment. In paper, the theory relative excitation test on tool expounded, carried then parameters are obtained experimental analysis. Analyze weak link center, show that stiffness Z direction feed component insufficient X, Y ram insufficient.

Existed conductor galloping can only analyze the real-time state. There is no effective method to predict state and alarm if necessary in future. In order solve above problems, a new based on MEMs 3Ddynamic displacement sensing network has been proposed realized. The dynamic sensor used sense distributed information. low-power wireless deployed transmit data back host quickly realize centralized treatment. three-degree freedom model built multi-dimensional horizontal, vertical torsional...

The long term evolution (LTE) based smart grid communication network is a promising candidate solution for the future power grid. In this paper, we discuss three channel estimation schemes LTE-based grids, including least square (LS), minimum mean error (MMSE), and discrete Fourier transform (DFT) estimators. simulations, compare these estimators, results show that DFT-based estimator can achieve best tradeoff between performance implementation complexity.

The electric demand of EV in the public transportation sector is increasingly important to future city’s power distribution and even infrastructure construction. According characteristic transportation, this paper analyzed influence factors load they were divided into three parts. Then a predicting model based on fuzzy clustering analysis method was put forward. We used BP (Back Propagation) neural network algorithm solve problem. Finally operated practical example. Results showed that could...