This paper proposes a load identification approach that combines Weighted Recursive Graph (WRG) and Convolutional Neural Networks (CNN) to meet the higher identification demands of residential users. Firstly, the single-period steady-state current signal is extracted from the aggregate signal; the current is decomposed into two forms: active current and reactive current by the Fryze power theory, and then the recursive matrix is used to convert the decomposed reactive current into a two-dimensional image data, and the CNN multi-label classifier is used to extract and learn WRG image features automatically. Finally, the convolutional neural network is used to extract and learn the features of WRG images to complete the task of multi-label load identification.

In February 2017, as the first multilateral agreement on trade since the establishment of the WTO, the Trade Facilitation Agreement (TFA) came into effect.Hitherto, however, there is still very little evidence on the factual effect of trade facilitation on promoting international trade.Based on panel data from 2017-2021, using a trade gravity model and the least squares regression analysis, this paper explores the impact of the implementation of the TFA on the import and export values between China and five RCEP countries (Indonesia, Malaysia, Singapore, Thailand, and Vietnam).The results demonstrate that China's trade facilitation level is significantly and positively related to the trade flow of goods.In addition, the population size of RCEP countries is also positively correlated with the trade flows between the two sides. 1 per cent of Trade Facilitation Index (TFI) can increase trade flows by US$10,352.9million.It is recommended to actively promote the development of trade facilitation, including the formation of RCEP logistics industry alliance, the establishment of one-stop digital customs, the acceleration of Cross-border E-commerce, as well as the reasonable use of blockchain.

With the extensive integration of the Internet, social networks and the internet of things, the social Internet of things has increasingly become a significant research issue. In the social internet of things application scenario, one of the greatest challenges is how to accurately recommend or match smart objects for users with massive resources. Although a variety of recommendation algorithms have been employed in this field, they ignore the massive text resources in the social internet of things, which can effectively improve the effect of recommendation. In this paper, a smart object recommendation approach named object recommendation based on topic learning and joint features is proposed. The proposed approach extracts and calculates topics and service relevant features of texts related to smart objects and introduces the “thing-thing” relationship information in the internet of things to improve the effect of recommendation. Experiments show that the proposed approach enables higher accuracy compared to the existing recommendation methods.

In order to realize on-line real-time monitoring of vehicle running status in enterprises, improve vehicle running supervision ability and vehicle mileage calculation accuracy, and improve vehicle compliance travel level, vehicle position positioning and track data collection are realized based on GPS positioning technology by installing vehicle intelligent terminal, and the positioning data are transmitted to the management platform through data communication service, real-time viewing and track playback of vehicle position and track are carried out through Siji map service through data receiving, processing and optimization processes. After filtering abnormal trajectory points, completing missing trajectory points and correcting trajectory, the vehicle trajectory can be fitted effectively, and the quality of vehicle trajectory data can be enhanced. Based on this, the purpose of vehicle mileage calculation optimization can be realized [1].

Based on the machinery industry characteristic background of school of mechanical and electrical engineering in Wuhan University of technology, combined with the use of industrial engineering experiment resources in Wuhan University of technology and related brother colleges, this paper analyzes the current situation and problems of experiment resource sharing, and studies the utilization efficiency of industrial engineering experiment resources at home and abroad. Using expert analysis, data mining, intelligent optimization operation and other methods, the reasons for this phenomenon were analyzed. Discusses on how to use programming language to develop and build a web-based industrial engineering experiment resource sharing management system. The following whole process management from the open experimental design, process control, experimental effect, experimental resource life and other processes of industrial engineering was carried out, in order to optimize the allocation of industrial engineering experiment resources sharing, and make full use of different experimental resources. Improved utilization rate of industrial engineering experimental resources would thus be obtained, so as to further adapt to the requirements of cultivating high-quality compound new engineering talents with strong engineering practice ability, strong innovation ability and international competitiveness.

With the development of remote sensing, aerospace, and other related technologies, remote sensing image application scenarios are more and more extensive. In this paper, the training of remote sensing data in different airports is improved based on the YOLOv5 (You Only Look Once) model. Based on the analysis of existing deep learning frameworks and models, a remote sensing image segmentation program is designed to solve the problems of large-scale, large variation, and multiple data channels of remote sensing images. To improve the performance of small object detection, the original image is segmented on the detector. When the size of the original image is large, it will be reduced in the detector input, and the number of pixel representatives of small targets will be seriously reduced, leading to the degradation of detector detection performance. So, we propose the purpose of image segmentation and separation detection is to reduce the loss of small target pixels during image reduction. It shows great advantages in solving the problem of precision and speed of aircraft identification, and it is feasible to identify aircraft with a computer quickly and efficiently.

This paper investigates the challenging fault prediction problem in process industries that adopt autonomous and intelligent cyber-physical systems (CPS), which is in line with the emerging developments of industrial internet of things (IIoT) and Industry 4.0. Particularly, we developed an end-to-end deep learning approach based on a large volume of real-time sensory data collected from a chemical plant equipped with wireless sensors. Firstly, a novel recursive architecture with multi-lookback inputs is proposed to perform autoregression on imbalanced time-series data as a preliminary prediction. In this process, a novel learning algorithm named recursive gradient descent (RGD) is developed for the proposed architecture to reduce cumulative prediction uncertainties. Subsequently, a classification model based on temporal convolutions over multiple channels with decay effect is proposed to perform multi-class classification for fault root cause identification and localization. The overall network is named the cumulative uncertainty reduction network (CURNet), for its superior capacity in reducing prediction uncertainties accumulated over multiple prediction steps. Performance evaluations show that CURNet is able to achieve superior performance especially in terms of fault prediction recall and fault type classification accuracy, compared to the existing techniques.

This work offers an enhanced intelligent picture text recognition algorithm based on the intelligent image text recognition method to increase the impact of English image text translation. Texture blocks of adaptable size are used to successfully increase the accuracy and efficiency of restoration due to the varying texture information present in various photos. Furthermore, the repair sequence is altered as a result of the improved priority calculation algorithm, and the weight of the structural information is enhanced at the same time. In addition, in order to reduce the overall structural complexity and calculation amount of the system, the gated loop unit is also selected as the RNN structure. Finally, this article constructs an English translation system based on intelligent image text recognition according to the requirements of intelligent image text recognition and designs experiments to evaluate the performance of the system constructed in this article. The experimental statistical results show that the English translation system constructed in this article can basically meet the needs of English image text recognition.

After the outbreak of COVID-19, the indoor environment has become particularly important in closed spaces, being a common concern in environmental science and public health, and of great significance for the building environment. To improve the indoor air quality and control the spread of viruses, the analysis of inhalable particles in indoor environments is critical. In this research, we study standards focused on inhalable particles and indoor environmental quality, as well as analyzing the movement and diffusion of indoor particles. Based on our analysis, we conduct an experimental study to determine the distribution of indoor inhalable particles of different sizes before and after diffusion under the conditions of underfloor air distribution. Furthermore, the mathematical modeling method is adopted to simulate the indoor flow field, particle trajectories, and pollutant dispersion process. The k-epsilon two-equation model is applied as the turbulence model in the numerical simulation, while the Lagrangian discrete phase model is adopted to trace the motion of particles and analyze the distribution characteristics of indoor particles. The results demonstrate that fine particles (i.e., those with size less than 0.5 mu m) have a significant impact on the indoor particle concentration, while coarse particles (i.e., with size above 2.5 mu m) have a greater influence on the total mass concentration of indoor particles. Small-sized particles can easily follow the airflow and diffuse to upper parts of the room. Overall, the effects of indoor particles on indoor air quality, including the potential threat of aerosol transmission of respiratory infectious diseases, are non-negligible. Application of the presented research can contribute to improving the health-related aspects of the building environment. (C) 2022 Elsevier B.V. All rights reserved.

•We introduce feature separation into traditional cross-modal retrieval task to deal with information asymmetry between different modalities, and use different loss functions to supervise different parts of the feature vectors.•We introduce image and text reconstruction tasks for specific information of images and texts, forcing the accuracy of feature separation operation and improving the quality of specific information.•We use the multi-task learning framework, integrate cross-modal retrieval tasks, image and text reconstruction tasks, and further improve the performance of cross-modal retrieval tasks through joint training.•We conduct extensive experimentation on MS-COCO and Flickr30K datasets. Our empirical results demonstrate that feature separation and specific information reconstruction can significantly improve the baseline performance of cross-modal image-text retrieval.

The Poggio-Miller-Chang-Harrington-Wu-Tsai (PMCHWT) equations are employed to describe the electromagnetic (EM) feature of homogeneous dielectric objects and extended to augmented PMCHWT (A-PMCHWT) equations for low-frequency EM modeling. The PMCHWT equations are much more widely-used than the individual electric field integral equations (EFIEs) since they can use the Rao-Wilton-Glisson (RWG) basis function to represent both electric and magnetic current densities in the method of moments (MoM) solution and the resultant impedance matrix is well-conditioned. At low frequencies, one has only augmented the EFIEs by using the dual basis function (DBF) to represent the magnetic current density. In this work, the PMCHWT equations are augmented by introducing the continuity equation of magnetic current density and magnetic charge density as the extra equation and unknown function and the complicated DBF can then be removed. Numerical examples are provided to demonstrate the effectiveness and robustness of the approach.

The surface integral equations (SIEs) are used to describe the electromagnetic (EM) problems with composite objects including both conducting and dielectric media. At low frequencies, the $\mathcal{L}$ operator in the SIEs tends to break down and the augmented electric field integral equation (AEFIE) was proposed for conducting objects. Later on, the AEFIE was extended to AEFIEs for dielectric objects, but the dual basis function (DBF) is needed to represent the magnetic current density in the method of moments (MoM) solutions. We consider the EM modeling for composite objects in this work and hybrid field integral equations (HFIEs) are proposed to govern the dielectric part. The HFIEs are augmented into AHFIEs which are combined together with the AEFIE of conducting part for the low-frequency modeling of the whole structure. A numerical example is presented to demonstrate the method which shows good performance.

Effectively reducing the line loss rate is a key task for power companies to improve enterprises economic efficiency. Among them, the substation bus bar power imbalance rate abnormality, user electricity metering anomaly, line-variable-table topological relationship anomaly, directly affects line loss rate statistical assessment and economics of power system operation. However, considering the large number of watt-hour meters, the types of faults, and the complicated wiring, it makes troubleshoot difficult to abnormal metering device. In order to quickly and accurately troubleshoot faults, this paper divides electricity consumption data into short-term data and medium-long-term data based on massive data according to the length of time scale. This paper also extracts electricity consumption data and feature quantity of user characterizing metering abnormality. Based on the differences in data characteristics, the user-side measurement anomaly screening methods at different time scales are targeted and designed.

移动机器人的足端轨迹规划是多维复杂难求解的问题,对此,文章提出了一种改进的果蝇优化算法(fruit fly optimization algorithm,FOA)对爬行机器人的该类问题进行设计优化.首先建立机器人关节空间足端轨迹模型,提出分段多项式曲线轨迹规划方法,同时将FOA引入自适应步长并与烟花算法爆炸操作混合,利用改进后的FOA对以时间最优为目标的足端轨迹算例进行优化验证.验证结果表明,该算法可行有效,优化结果优于标准的FOA.

A new solution for the remote control system in smart home is introduced. With Bluetooth wireless communication technology, this system realizes convenient and flexible home-networking. The remote control for network appliances through telephones or Internet is achieved,and the centralized control for home-networked equipments based on wirless endpoints also implemented. This system is simple to use and easy to be extended.

Electromagnetic interaction with elastic objects includes both electrodynamic and elastodynamic processes which are of multiphysics feature. The electrodynamic process is governed through Maxwell's equations while the elastodynamic process is described by elastic wave equations and they are coupled together by excitations. Traditionally, these equations are presented in the form of partial differential equations and only differential equation solvers can be used to solved them. In this work, we develop coupled integral equations for the multiphysics process based on equivalence theorem so that integral equation solvers can be used to solve them. A typical numerical example is presented to demonstrate the approach and good results have been obtained.

Electromagnetic problems with both conducting media and penetrable media are formulated by volume-surface integral equations (VSIEs) in the integral equation approach. Conventionally, the electric field integral equation (EFIE) is used to describe the conducting part, but it has a serious low-frequency breakdown problem. In this work, we use the magnetic field integral equation to replace the EFIE for the conducting part and combine the volume integral equations of the penetrable part to form the VSIEs. The resulting VSIEs belong to a second kind of integral equations and can remove the low-frequency breakdown problem. A numerical example is presented to demonstrate the approach and good results have been obtained.

Al-5Mg-0.8Mn alloys (AA5083) with various iron and silicon contents were cast under near-rapid cooling and rolled into sheets. The aim was to study the feasibility of minimizing the deteriorating level of the harmful Fe-rich phases on the mechanical properties through refining the intermetallics by significantly increasing the casting rate. The results showed that the size and density of the intermetallic particles that remained in the hot bands and the cold rolled sheets increased as the contents of iron and silicon in the alloys were increased. However, the increment of the particle sizes was limited due to the significant refinement of the intermetallics formed during casting under near-rapid cooling. The mechanical properties of the alloys reduced as the contents of iron and silicon in the alloys increased. However, the decrement of tensile strengths and ductility was quite small. Therefore, higher contents of iron and silicon could be used in the Al-5Mg-0.8Mn alloy (AA5083 alloy) when the material is cast under near-rapid cooling, such as in the continuous strip casting process.

An Al-based composite reinforced with core–shell-structured Ti/Al3Ti was fabricated through a powder metallurgy route followed by hot extrusion and was found to exhibit promising mechanical properties. The ultimate tensile strength and elongation of the composite sintered at 620°C for 5 h and extruded at a mass ratio of 12.75:1 reached 304 MPa and 14%, respectively, and its compressive deformation reached 60%. The promising mechanical properties are due to the core–shell-structured reinforcement, which is mainly composed of Al3Ti and Ti and is bonded strongly with the Al matrix, and to the reduced crack sensitivity of Al3Ti. The refined grains after hot extrusion also contribute to the mechanical properties of this composite. The mechanical properties might be further improved through regulating the relative thickness of Al–Ti intermetallics and Ti metal layers by adjusting the sintering time and the subsequent extrusion process.

Power battery equalization is an important technology in the application of electric vehicles.At present, the balance degree of battery pack is the only control criterion in the majority of equalizer circuit and control strategies.The energy utilization rate is ignored.In this paper, a balanced circuit combining the distributed-load balancing and the central-load balancing is designed to improve the balancing speed to a large extent.An equilibrium control strategy based on transferring least energy is also put forward.The results of simulation experiment show that the equalization circuit is correct.The control strategy can improve the energy utilization and reduce the loss.This method can be used in practical applications.