LUT_IR
兰州理工大学
http:///ir.lut.edu.cn:80
2024-03-19T08:36:51Z
2024-03-19T08:36:51Z
Communication Base Station Traffic Prediction Model Based on EMD-RF-Transformer
Liu, Qinhe
Wang, Senlin
Yang, Longrui
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169680
2024-02-23T08:25:46Z
2024-02-23T08:25:45Z
题名: Communication Base Station Traffic Prediction Model Based on EMD-RF-Transformer
作者: Liu, Qinhe; Wang, Senlin; Yang, Longrui
摘要: In this article, an innovative communication base station traffic prediction model is proposed for efficiently and accurately predicting traffic data. The model combines empirical mode decomposition (EMD), Transformer neural network, and random forest feature importance (RF-FI) methods. This article provides a reliable prediction model for the construction and maintenance of future communication network base stations, which is expected to have significant practical applications. Firstly, the EMD method decomposes the time series signal of the base station into feature quantities of different frequencies. Then, the importance of features is evaluated using the RF-FI method. The feature with the highest score is selected and trained on the Transformer model to predict the new sequence. Experimental results demonstrate a significant increase in forecasting accuracy (87%) and better overall performance compared to single use of the Transformer model and other time series forecasting models such as the LSTM model. Overall, this communication base station traffic prediction model is an effective and reliable solution for predicting traffic data, providing practical approaches to base station construction and maintenance, and valuable insights for future research. © 2023 IEEE.
2024-02-23T08:25:45Z
Research on weakening control strategy of permanent magnet synchronous motor of high speed vacuum cleaner based on sliding mode observer
Xu, Mengzhuo
Qiu, Ziheng
Wang, Zhenhua
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169679
2024-02-23T08:25:44Z
2024-02-23T08:25:44Z
题名: Research on weakening control strategy of permanent magnet synchronous motor of high speed vacuum cleaner based on sliding mode observer
作者: Xu, Mengzhuo; Qiu, Ziheng; Wang, Zhenhua
摘要: The built-in permanent magnet synchronous motor (IPMSM) has been widely used in electric vehicles, aerospace, national defense, robotics, medical and other fields due to its high operating efficiency, high torque density and relatively stable running condition. Traditional IPMSM control strategy includes speed control of position sensor, magnetic field orientation control and so on. They have the limitations of increasing system cost and algorithm failure in case of sensor failure. In this paper, high-speed vacuum cleaner as the research object, aiming at the vacuum cleaner in high speed, sensorless drive motor high-speed operation ability, using sliding mode control and weak magnetic control strategy. A series of experiments and simulations are designed. The results show that the proposed sliding mode and magnetic weakening control strategies have good dynamic response and steady state response, and more accurate estimated position and estimated speed are obtained. The system has the advantages of strong robustness. © 2023 IEEE.
2024-02-23T08:25:44Z
IUCNet: A Novel Semantic Segmentation Network for Overlapping Industrial Parts Detection
Wu, Kedi
Yang, Qingcheng
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169678
2024-02-23T08:25:44Z
2024-02-23T08:25:44Z
题名: IUCNet: A Novel Semantic Segmentation Network for Overlapping Industrial Parts Detection
作者: Wu, Kedi; Yang, Qingcheng
摘要: General-purpose segmentation methods networks are difficult to adapt to more complex industrial scenes based on convolutional neural, such as industrial parts with more stacking, tilting, and occlusion cases. This paper's 2D image dataset with multiple scene occlusions is constructed using two common types of parts, bolts, and shims. Based on this, a semantic segmentation method, IUCNet, is validated with excellent and stable results. This method selects UperNet as the model, ConvNeX as the backbone network, and the combination of Dice Loss and Cross Entropy as the loss function, based on the excellent performance in other segmentation tasks. Meanwhile, we also introduce a photometric distortion pre-processing module to solve the blurring of overlapping edges caused by shadows and similar colors in industrial part images. Our method finally obtains an average IoU of 91.32 and an average PA of 96.17 in the self-built image test set. © 2023 IEEE.
2024-02-23T08:25:44Z
Model Predictive Current Control of Permanent Magnet Synchronous Motor Based on Optimized Extended Voltage Vectors
Meng, Qingcheng
Bao, Guangqing
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169677
2024-02-23T08:25:43Z
2024-02-23T08:25:43Z
题名: Model Predictive Current Control of Permanent Magnet Synchronous Motor Based on Optimized Extended Voltage Vectors
作者: Meng, Qingcheng; Bao, Guangqing
摘要: In the conventional single vector model predictive current control and duty model predictive current control, because only six effective voltage vectors in the fixed directions can be selected in each sampling period, there is a large current ripple. To solve this problem, a model predictive current control method is proposed to optimize the extended voltage vectors. The method can select the optimal voltage vector from 14 alternative voltage vectors to minimize the cost function, which makes the selection of voltage vector more accurate. The simulation results show that the proposed method has good dynamic and static performance. Compared with the conventional single vector or duty-cycle model predictive current control method, the proposed method effectively reduces the current ripple and improve the current quality. © 2023 IEEE.
2024-02-23T08:25:43Z
Study on effect of injection temperature on shear strength of the direct bond between polyphenylene sulfide and anodized aluminum alloy
Zhao, Wei
Qiu, Jianhui
Sakai, Eiichi
Wu, Haonan
Feng, Huixia
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169676
2024-02-23T08:25:42Z
2024-02-23T08:25:42Z
题名: Study on effect of injection temperature on shear strength of the direct bond between polyphenylene sulfide and anodized aluminum alloy
作者: Zhao, Wei; Qiu, Jianhui; Sakai, Eiichi; Wu, Haonan; Feng, Huixia
摘要: A simple and environmentally friendly technique has been proposed to form a composite structure of high interfacial bonding strength between polyphenylene sulfide (PPS) and aluminum alloy (Al). The Al substrate is modified through anodization and enlarging treatments. The influences of injection temperature on the anodic aluminum oxide film surface on the interfacial bond strength of the formed PPS-Al is discussed. The results revealed that when the injection temperature of PPS was set at 310°C, the bonding strength increased to 18.0 MPa. The adhesive strength exhibits a clear dependency on the injection temperature. At higher injection temperatures, PPS exhibits higher flowability and better penetration into nanoscale pores, resulting in a strong interlocking between PPS and anodized aluminum alloy. © Published under licence by IOP Publishing Ltd.
2024-02-23T08:25:42Z
Influence of Perceived Authenticity of Virtual Spokesperson on Consumer Brand Attitude
Hu, Guangkuo
Zhao, Yanyan
Samia, Elghazali
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169675
2024-02-23T08:25:41Z
2024-02-23T08:25:41Z
题名: Influence of Perceived Authenticity of Virtual Spokesperson on Consumer Brand Attitude
作者: Hu, Guangkuo; Zhao, Yanyan; Samia, Elghazali
摘要: With the rapid development of digital technologies such as virtual reality (VR), artificial intelligence (AI) and computer-generated image (CGI), the types of virtual spokespersons are more and more abundant and widely applied. Virtual spokespersons build a new communication channel between brands and consumers, so it is urgent for enterprises to explore the influence of perceived authenticity of virtual spokespersons on consumers' brand attitudes. Based on the perspective of psychological distance, through the empirical analysis of the results, it is found that the perceived authenticity of virtual spokesperson has a significant positive impact on psychological distance and brand attitude, and psychological distance has a significant positive impact on brand attitude. Psychological distance plays a partial mediating role between virtual spokesperson's perceived authenticity and brand attachment. © 2023 IEEE.
2024-02-23T08:25:41Z
Simulation study of transient overvoltage at HVDC transmission terminal
Lu, Zhiyuan
Jin, Hai
Zhang, Hongliang
Wang, Hong
Li, Xiang
Ma, Zhiying
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169674
2024-02-23T08:25:40Z
2024-02-23T08:25:39Z
题名: Simulation study of transient overvoltage at HVDC transmission terminal
作者: Lu, Zhiyuan; Jin, Hai; Zhang, Hongliang; Wang, Hong; Li, Xiang; Ma, Zhiying
摘要: This For high voltage DC transmission system, DC phase change failure shall cause transient overvoltage at the transmission side of the grid bus, and between the delivery system and the DC system the reactive power exchanged will change dramatically when the phase change fails. Therefore, we analyze the mechanism and characteristics of the transient overvoltage at the AC bus at the transmission end after the phase change failure caused by the AC fault on the receiving end, and build a system model of positive and negative 800KV high-voltage DC transmission Using the PSCAD-EMTDC simulation platform, an analysis is conducted to examine the transient overvoltage and reactive power variations at the transmitter bus during different AC faults. Additionally, the transient behavior of the DC system voltage and current is studied. Through adjusting and optimizing the parameters of the fixed control system. Obtained through simulation that when Phase change fault occurs, the transient characteristics of the sending bus voltage exhibit a decreasing trend followed by an increase. During a phase change failure, the reactive power flow between the receiving system and the DC system is reversed. The DC system acts as a reactive power absorber, drawing in a considerable amount of reactive power from the sending system. Subsequently, it releases a substantial imaginary power back into the sending system. ©2023 IEEE.
2024-02-23T08:25:39Z
A MIMO Detector Utilizing DNN for Faster-Than-Nyquist Optical Wireless Communications
Cao, Minghua
Yao, Ruifang
Sun, Qinxue
Zhang, Yue
Yang, Qing
Wang, Huiqin
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169673
2024-02-23T08:25:39Z
2024-02-23T08:25:38Z
题名: A MIMO Detector Utilizing DNN for Faster-Than-Nyquist Optical Wireless Communications
作者: Cao, Minghua; Yao, Ruifang; Sun, Qinxue; Zhang, Yue; Yang, Qing; Wang, Huiqin
摘要: Traditional multiple input multiple output (MIMO) detection algorithms encounter challenges with computational complexity and performance limitations when dealing with high-dimensional inputs and complex channel conditions. To improve the precision of signal recovery in atmospheric turbulence channels for faster-than-Nyquist (FTN) optical wireless communication (OWC) systems, a deep learning (DL) based MIMO detector is proposed. By utilizing a deep neural network (DNN), nonlinear mappings in MIMO systems can be learned, leading to improved detection performance and reduced computational overhead. Simulation results demonstrate that our proposed DNN detector can achieve comparable performance to the maximum likelihood (ML) method and the complexity is reduced by 37% compared with the ML. © 2023 IEEE.
2024-02-23T08:25:38Z
Misaligned bilateral linear permanent magnet vernier motor design and optimization
Yang, Qiaoling
Zhang, Qi
Su, Binhua
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169672
2024-02-23T08:25:37Z
2024-02-23T08:25:37Z
题名: Misaligned bilateral linear permanent magnet vernier motor design and optimization
作者: Yang, Qiaoling; Zhang, Qi; Su, Binhua
摘要: Power factor analysis is crucial while researching linear permanent magnet vernier motors. In this study, a new staggered structure motor is created by adopting the technique of staggered armature teeth and windings to implement the magnetic circuit series connection, in accordance with the features of magnetic circuit series connection to improve the effective magnetic flux. It efficiently lowers magnetic leakage and raises the motor's power factor. Combining the finite element method with the analytical method, firstly, the magnetic circuit analysis is carried out, and secondly, based on this, the effects of individual parameters on the average thrust and thrust pulsation of the motor are analyzed with the optimization objectives of increasing the average thrust and decreasing the thrust pulsation of the motor. Finally, the finite element calculation results of symmetric and staggered motors are compared, and the simulation results show that the staggered motor thrust is improved by 80.71N and the thrust pulsation is reduced by 12.47%, which proves the effectiveness of the proposed method. © 2023 IEEE.
2024-02-23T08:25:37Z
Case Study of Ant Group's "double Link" Financing Based on Blockchain Technology
Song, Lili
http:///ir.lut.edu.cn:80/handle/2XXMBERH/169671
2024-02-23T08:25:36Z
2024-02-23T08:25:36Z
题名: Case Study of Ant Group's "double Link" Financing Based on Blockchain Technology
作者: Song, Lili
摘要: Through the investigation of the representative enterprises of supply chain finance, the operation and service mode, development status and development bottleneck of supply chain finance are analyzed. The "blockchain+supply chain financing" mode of Ant Group is discussed at the theoretical level by taking the "double link" financing mode as an example, and the underlying framework is built with the help of blockchain technology. Enable supply chain finance service providers to establish a comprehensive service platform for all participants in the supply chain, and solve the difficulties of supply chain finance in supporting the financing of small and medium-sized enterprises through the combined application of multiple functional modules, sum up experience, and put forward optimization ideas and strategies. In order to enrich the research content of supply chain financing optimization and innovation, and improve the theoretical system of "blockchain+supply chain financing". © 2023 Elsevier B.V.. All rights reserved.
2024-02-23T08:25:36Z