Patents Assigned to Hefei University of Technology
-
Publication number: 20220074027Abstract: A high-hardness composite oxide dispersion-strengthened tungsten alloy and a preparation method thereof are disclosed. The high-hardness composite oxide dispersion-strengthened tungsten alloy consists essentially of a tungsten phase, and nano-scale Y2O3 and ZrO2 particles dispersed in the tungsten phase, wherein there is a Y—Zr—O ternary phase structure at a coherent/semi-coherent interface.Type: ApplicationFiled: July 13, 2021Publication date: March 10, 2022Applicant: Hefei University of TechnologyInventors: Laima LUO, Zhihao Zhao, Yucheng Wu, Gang Yao, Xiang Zan, Xiaoyong Zhu
-
Publication number: 20220075203Abstract: The present disclosure provides a light field display device, including: a backlight layer; a light control array layer; and a control layer. The backlight layer, the light control array layer, and the control layer are arranged side by side, and images on the backlight layer, the light control array layer, and the control layer are superimposed to form a three-dimensional image.Type: ApplicationFiled: November 13, 2019Publication date: March 10, 2022Applicant: Hefei University of TechnologyInventors: Guoqiang LV, Liming ZHU, Lei WU, Jiechao SHENG, Qibin FENG, Zi WANG
-
Publication number: 20210387351Abstract: The present disclosure relates to a rope traction type grinding, cleaning, and coating integrated operation robot. The operation robot includes a hanging basket, a first traction mechanism connected to the hanging basket, a grinding mechanism arranged in front of the hanging basket, and a cleaning and spraying mechanism and a spring reaction force regulation mechanism arranged in the hanging basket. The first traction mechanism includes first ropes for connecting the hanging basket and first rope winding mechanisms. The cleaning and spraying mechanism includes a first vertical plate and a second vertical plate that are arranged in parallel in a vertical direction. A cleaning nozzle and a spraying nozzle are mounted on the first vertical plate. From the above technical solution, it can be seen that the operation robot adopts a rope traction manner, and has the advantages of large work space, low mechanism inertia, and accurate and reliable location.Type: ApplicationFiled: January 7, 2021Publication date: December 16, 2021Applicant: Hefei University of TechnologyInventors: Daoming WANG, Pengtao XIE, Bin ZI, Zhengyu WANG, Sen QIAN, Zitong HUANG
-
Patent number: 11195051Abstract: The invention relates to a method for person re-identification based on deep model with multi-loss fusion training strategy. The method uses a deep learning technology to perform preprocessing operations such as flipping, clipping, random erasing and style transfer, and then feature extraction is performed through a backbone network model; joint training of a network is performed by fusing a plurality of loss functions. Compared with other deep learning-based person re-identification algorithms, the present invention greatly improves the performance of person re-identification by adopting a plurality of preprocessing modes, the fusion of three loss functions and effective training strategy.Type: GrantFiled: March 9, 2020Date of Patent: December 7, 2021Assignees: Tongji University, Hefei University of Technology, Beijing E-Hualu Info Technology Co., Ltd.Inventors: Deshuang Huang, Sijia Zheng, Zhongqiu Zhao, Xinyong Zhao, Jianhong Sun, Yang Zhao, Yongjun Lin
-
Patent number: 11181884Abstract: According to a machining principle of the CNC gear hobbing machine, a functional relation between a geometric error of a gear and a tracking error of each motion axis of the machine tool is constructed; a machining error mathematical model of tooth profile deviation, tooth pitch deviation and tooth direction deviation at each position control time point is established by tracking errors of each motion axis; a compensation quantity required for a workpiece rotation axis at the next position control time point is calculated by establishing a decoupling compensation model; average absolute values of machining errors and a total compensation quantity of the machining errors under the conditions of not adopting the synchronous control method and adopting the synchronous control method in the total position control time are obtained by calculating machining error values of each position controls time point, and the synchronized multi-axis motion control is completed.Type: GrantFiled: April 21, 2020Date of Patent: November 23, 2021Assignee: Hefei University of TechnologyInventors: Xiaoqing Tian, Dan Li, Yu Wu
-
Publication number: 20210334333Abstract: The present application discloses a verification system and method for an operation result based on a reconfigurable butterfly unit. The system is applicable to a digital signal processing (DSP) chip. The DSP chip includes a reconfigurable butterfly unit. The reconfigurable butterfly unit may be reconfigured into two modes: a first verification mode and a second verification mode. The system includes: a controller, a memory, a verification unit, a first data gating unit, and a second data gating unit. The technical solution in the present application is used to overcome the disadvantage that an existing verification system and an existing verification method consume large hardware resources, thereby reducing the implementation costs of operation result verification.Type: ApplicationFiled: February 10, 2021Publication date: October 28, 2021Applicant: HeFei University of TechnologyInventors: Yukun Song, Zhengmao Wang, Duoli Zhang, Xu Tang, Zhenmin Li, Gaoming Du
-
Publication number: 20210299860Abstract: The present invention provides a method for robot action imitation learning in a three-dimensional space and a system thereof, relates to the technical fields of artificial intelligence and robots. A method based on a series-parallel multi-layer backpropagation (BP) neural network is designed for robot action imitation learning in a three-dimensional space, which applies an imitation learning mechanism to a robot learning system, under the framework of the imitation learning mechanism, to train and learn by transmitting demonstrative information generated from a mechanical arm to the series-parallel multi-layer BP neural network representing a motion strategy.Type: ApplicationFiled: October 13, 2020Publication date: September 30, 2021Applicant: Hefei University of TechnologyInventors: Minglun REN, Yuanyuan MA
-
Patent number: 11116682Abstract: A medical apparatus for standing aid includes a backrest (4) that has at least one side thereof provided with a crank rocker mechanism that includes a crank mechanism (1), a triangle linkage mechanism (2), and a rocker mechanism (3), the crank mechanism (1) is rotatably connected to the triangle linkage mechanism (2) such that an included angle between a second driven link (BE) of the triangle linkage mechanism (2) and the crank mechanism (1) is always smaller than 90°, and the crank mechanism (1) driving the triangle linkage mechanism (2), the rocker mechanism (3) and the backrest (4) connected to the triangle linkage mechanism to perform interactive movement repeatedly along a predetermined curve trajectory in response to a driving force from a drive effect of a driving unit (100), so that the backrest (4) connected to the second driven link (BE) assists a trainee in standing up repeatedly by obliquely supporting the trainee's waist.Type: GrantFiled: August 30, 2018Date of Patent: September 14, 2021Assignee: Hefei University of TechnologyInventors: Ning An, Liuqi Jin
-
Patent number: 11067992Abstract: The present invention provides a path planning method and system for self-driving of autonomous system, and relates to the technical field of autonomous systems. The method comprises following steps of: acquiring a path optimization function of an agent; converting, based on fixed-point theorems, the path optimization function of the agent into an equivalent fixed-point equation; acquiring a complete simplex sequence based on the fixed-point equation; and, determining, based on the complete simplex sequence, an initial population size and an initial position of particles for particle swarm optimization to obtain the best path planning of the agent. In the present invention, the extremal optimization of the path optimization function of the agent is converted into solving of the fixed-point equations, and initial parameters for particle swarm optimization are determined by the complete simplex sequence.Type: GrantFiled: March 18, 2020Date of Patent: July 20, 2021Assignee: Hefei University of TechnologyInventors: Minglun Ren, Xiaodi Huang, Chenze Wang, Bayi Cheng
-
Patent number: 11059936Abstract: The present invention discloses an organofluorine modified epoxy resin for increasing the surface flashover voltage, and preparation methods and applications thereof. In the present invention, the organofluorine modified epoxy resin is prepared from bisphenol A epoxy resin and hexafluorobisphenol A as raw materials in a mild environment in the presence of tri-(dimethylaminomethyl)phenol as a catalyst. Due to the reasonable proportion of the raw materials, the performance of the resin is optimized, the process is simple, and it is easy to realize industrial production. By using the modified resin as a matrix and adding a curing agent, a filler and an accelerant, a modified resin composite is prepared. In the present invention, a —CF3 group is introduced in the epoxy resin without changing the surface state of the epoxy composite insulating material, the volume/surface resistivity of the material is reduced, it is less likely to accumulate charges on the material surface.Type: GrantFiled: October 30, 2019Date of Patent: July 13, 2021Assignee: Hefei University of TechnologyInventors: Yushun Zhao, Kerong Yang, Jinhui Bai, Xuepei Wang, Yuanhan He
-
Patent number: 11062796Abstract: The present invention discloses a multimode mobile electronic medical record system and a working method thereof. The multimode mobile electronic medical record system comprises a plurality of mobile terminals, service server, push server, authentication server and cloud server. The mobile terminal comprises a medical record information collection module, a medical record generation module, a medical record synchronization module, a medical record parsing module and a medical record showing module. The service server comprises a medical record storage module and a medical record exchange module. The push server comprises a medical record push module. The authentication server comprises a medical record safety control module. The multimode mobile electronic medical record system can meet collection, integration and transfer of multimode electronic medical record information in a mobile medical environment, and can efficiently improve efficiency and convenience of mobile medical services.Type: GrantFiled: November 3, 2017Date of Patent: July 13, 2021Assignee: Hefei University of TechnologyInventors: Shuai Ding, Shanlin Yang, Wenjuan Fan, Feng Niu, Zeyuan Wang, Ling Li, Zijie Yue
-
Publication number: 20210114385Abstract: The present invention discloses a method for making an easy-to-view computed tomography (CT) sheet. The method is implemented by the following modules, including an X-ray tomographic scanning device, a computer system, an operating system, and a printing system. The X-ray tomographic scanning device is composed of an X-ray generator, an X-ray tube, and an X-ray detector. The X-ray generator of the X-ray tomographic scanning device is used for X-ray generation. Through increasing detection accuracy of the detector inside the scanning device, radiation dose can be effectively reduced, thereby avoiding large radiation hazards to a patient during scanning. In this way, image quality is improved without increasing the radiation dose or even when the radiation dose is reduced. In addition, a processing method for image reconstruction is further improved under joint action of multiple groups of X-rays, thereby further improving data generation.Type: ApplicationFiled: October 20, 2020Publication date: April 22, 2021Applicant: Hefei University of TechnologyInventors: Xiaojia Wang, Kuo Du, Keyu Zhu, Sheng Xu
-
Patent number: 10836231Abstract: The present invention discloses an automatic adjustment device for a front suspension of a driverless formula racing car, comprising: a vertical column, an upper fork arm, a lower fork arm, a strength adjuster, a push rod, a rocking block and a shock absorber. By arranging the strength adjuster between the upper fork arm and the lower fork arm, and by adjusting the degree of tightness between the upper fork arm and the lower fork arm, the present invention can automatically adjust the tensile and support strength of the strength adjuster according to different car conditions and road conditions in which the racing car runs, to guarantee the structure stability of the front suspension in the running process, and prevent the front suspension from being loosening or fracturing, thereby having good adaptability and stability to structural deformation.Type: GrantFiled: March 29, 2019Date of Patent: November 17, 2020Assignee: Hefei University of TechnologyInventors: Yongjia Wang, Bofu Wu, Shuhua Li, He Huang, Wuhua Jiang, Ruifeng Li, Jingting Yang, Junzhao Yu
-
Publication number: 20200285896Abstract: The invention relates to a method for person re-identification based on deep model with multi-loss fusion training strategy. The method uses a deep learning technology to perform preprocessing operations such as flipping, clipping, random erasing and style transfer, and then feature extraction is performed through a backbone network model; joint training of a network is performed by fusing a plurality of loss functions. Compared with other deep learning-based person re-identification algorithms, the present invention greatly improves the performance of person re-identification by adopting a plurality of preprocessing modes, the fusion of three loss functions and effective training strategy.Type: ApplicationFiled: March 9, 2020Publication date: September 10, 2020Applicants: Tongji University, Hefei University of Technology, Beijing E-Hualu Info Technology Co.,Ltd.Inventors: Deshuang HUANG, Sijia ZHENG, Zhongqiu ZHAO, Xinyong ZHAO, Jianhong SUN, Yang ZHAO, Yongjun LIN
-
Publication number: 20200209892Abstract: The embodiments of the present disclosure provide a method and system for patrolling an expressway by unmanned aerial vehicles (UAVs). The method is applied to the system including a control center, a plurality of UAVs and a plurality of advertising board charging base stations. The method firstly receives unmanned aerial vehicle (UAV) status information sent by the plurality of unmanned aerial vehicles and road section patrolling data through the control center, and receives advertising board charging base station information sent by the plurality of advertising board charging base stations, then classifies, stores and analyzes the received data and information, and finally generates and sends control commands for the UAVs and charging advertising boards.Type: ApplicationFiled: November 19, 2019Publication date: July 2, 2020Applicant: Hefei University of TechnologyInventors: He LUO, Peng ZHANG, Guoqiang WANG, Moning ZHU, Xiaoxuan HU, Ju WANG, Xiaoduo LI, Wei XIA, Peng JIN, Huawei MA
-
Patent number: 10664494Abstract: The present invention provides a method and system for synchronously storing multi-modal information of portable endoscope. The method includes: S1: when a mobile terminal acquires data collected by a portable endoscope and updates local data with the acquired data, acquiring update record of the local data of the mobile terminal; and, S2: when the mobile terminal is synchronized with a cloud, synchronizing data in the mobile terminal with data in the cloud by using the update record as information to be synchronized. By creating a table to be synchronized to store change information and location information of files and selectively performing synchronization from the mobile terminal to the cloud or from the cloud to the mobile terminal according to whether or not the table to be synchronized is null during synchronization, unnecessary synchronization operations, storage burden of the mobile terminal and synchronization time and traffic are all reduced.Type: GrantFiled: July 30, 2018Date of Patent: May 26, 2020Assignee: Hefei University of TechnologyInventors: Shuai Ding, Shanlin Yang, Qiang Fu, Lin Zhang
-
Patent number: 10571871Abstract: The present invention discloses an automatic loading system and method for service functions of a hydraulic machine. The loading system includes an acquiring portion for acquiring i service function component chains of a hydraulic machine, a determining portion for determining a demand sequence of the service function component chains used by a user within each preset time period in a measuring cycle based on the acquired service function component chains and a deciding portion for automatically loading a service function of the hydraulic machine in a next moment according to the demand sequence based on a current input operation command of the user. By adopting the automatic loading system or method provided by the present invention, response speed of intelligent services of the hydraulic machine can be improved.Type: GrantFiled: August 6, 2018Date of Patent: February 25, 2020Assignee: Hefei University of TechnologyInventors: Qiang Zhang, Shanlin Yang, Zhanglin Peng, Jun Pei, Xiaonong Lu
-
Publication number: 20200012704Abstract: The application discloses a data processing device, a data processing method and a digital signal processing device. The data processing device is used to reduce computing amount by reading and writing operation of memory. The data processing device comprises: a module for calculating reduced coefficient matrices, a storage module, a module for modifying reduced coefficient matrices, a module for triangular inversing, a module for obtaining inversion matrices and a module for correcting reverse result.Type: ApplicationFiled: July 3, 2019Publication date: January 9, 2020Applicant: HeFei University of TechnologyInventors: Duoli Zhang, Ziyan Ye, Qi Sun, Yukun Song, Gaoming Du
-
Patent number: 10409639Abstract: The present invention discloses a task scheduling system and method suitable for mobile health, the characteristics thereof including: a mobile health task distribution module, a plurality of mobile health task processing modules and a mobile health task callback module; the mobile health task distribution module comprises: a resource description unit, a resource scheduling unit, a resource matcher; each of the mobile health task processing modules comprises: a task decomposition unit, a task execution unit, a task evaluation unit. The present invention is capable of carrying out tasks with full use of the currently available task processing resources, thereby being capable of improving the efficiency and performance of the task scheduling processing in the mobile health environment.Type: GrantFiled: November 10, 2017Date of Patent: September 10, 2019Assignee: Hefei University of TechnologyInventors: Shanlin Yang, Shuai Ding, Wenjuan Fan, Hao Wang, Jinxin Pan, Shikang Hu, Hui Huang
-
Patent number: 10402404Abstract: The present invention discloses a scheduling method and system based on a hybrid variable neighborhood search and gravitational search algorithm. The method includes: 1 setting parameters of the algorithm; 2 initializing an initial solution of the algorithm; 3 performing local search based on a gravitational search algorithm (GSA); 4 updating the initial solution; 5 determining whether an algorithm termination condition is satisfied; if yes, outputting the global optimal solution searched for by the algorithm, otherwise, returning to the step 3. According to the present invention, a near-optimal solution for the continuous batch processing problem based on position learning effect and linear starting time can be obtained, so that an enterprise can make full use of production resources thereof to the utmost extent, and thus reduce production costs and improve the enterprise service level and the customer satisfaction level.Type: GrantFiled: September 11, 2018Date of Patent: September 3, 2019Assignee: Hefei University of TechnologyInventors: Jun Pei, Wenjuan Fan, Xinbao Liu, Qiang Zhang, Min Kong, Shaojun Lu