当前查询到68条专利与查询词 "Zheng Yuhui"相关,搜索用时0.5000129秒!排序方式:
发明专利:36实用新型: 28外观设计: 4
36 条,当前第 1-10 条 下一页 最后一页 返回搜索页
申请号:201911398110.7 公开号:CN111131487A 主分类号:H04L29/08
摘要:【中文】本申请公开了一种深度学习平台容器镜像获取、共享方法及系统,应用于主节点,本申请仅在全部节点未存储有与训练任务请求对应的目标容器镜像的时候,才由主节点直接从容器镜像库中获取目标容器镜像,尽可能的将目标容器镜像的下载过程转换为利用集群系统内部网络进行节点之间互传的方式提高目标容器镜像的下载过程,同时,节点之间采用点对点的传输方式传输目标容器镜像,进一步的提高节点之间容器镜像的传输速度,且原本未存储目标容器镜像的第二节点得到目标容器镜像后,可以变为第一节点一同进行目标容器镜像的传输,加快目标容器镜像的传输速度,为训练任务能够尽快执行提供了前提,提高了训练效率。 【EN】The application discloses a method and a system for acquiring and sharing container mirror images of a deep learning platform, which are applied to a master node, wherein the master node directly acquires the target container mirror images from a container mirror image library only when all nodes do not store the target container mirror images corresponding to a training task request, the downloading process of the target container mirror images is converted into a mode of mutual transmission between the nodes by using an internal network of a cluster system as much as possible to improve the downloading process of the target container mirror images, meanwhile, the target container mirror images are transmitted between the nodes in a point-to-point transmission mode, the transmission speed of the container mirror images between the nodes is further improved, and after a second node which does not store the target container mirror images obtains the target container mirror images, the second node can be changed into a first node to transmit the target container mirror images together, the transmission speed of the target container mirror images is accelerated, and a premise is provided for the training task to be, the training efficiency is improved.
详细信息 下载全文

申请号:201911242263.2 公开号:CN110916378A 主分类号:A47B81/00
申请人:【中文】厦门大学【EN】XIAMEN University 申请日:2019.12.06 公开日:2020.03.27
摘要:【中文】本发明涉及计算机设备技术领域,具体是一种方便存放计算机设备的装置,包括壳体,壳体内侧固定连接设置有横板,横板顶部滑动连接设置有用于存放计算机的放置板,放置板上侧设置有与壳体固定连接的固定机构,放置板右侧设置有箱门,放置板左侧与壳体之间固定连接设置有第二伸缩杆,壳体底部设置有支撑脚,本发明,通过设置可以自动移动的放置板,方便人们存放和拿取计算机,省时省力,通过设置风机和气囊,可以通过向气囊充气使计算机被气囊和挡板固定,且通过控制充气量,可以使装置可以对不同大小的计算机进行固定,通过设置固定机构,可以对计算机提供进一步的固定,有效保证了计算机存放时固定的稳定性。 【EN】The invention relates to the technical field of computer equipment, in particular to a device convenient for storing computer equipment, which comprises a shell, wherein the inner side of the shell is fixedly connected and provided with a transverse plate, the top of the transverse plate is slidably connected and provided with a placing plate for storing a computer, the upper side of the placing plate is provided with a fixing mechanism fixedly connected with the shell, the right side of the placing plate is provided with a box door, the left side of the placing plate is fixedly connected and provided with a second telescopic rod, and the bottom of the shell is provided with a supporting leg. The stability of fixing when the computer is deposited has effectively been guaranteed.
详细信息 下载全文

申请号:201880046172.9 公开号:CN110870283A 主分类号:H04L29/06
申请人:【中文】PQ解决方案有限公司【EN】PQ Solutions Limited 申请日:2018.04.25 公开日:2020.03.06
摘要:【中文】描述了用于用户身份和交易认证的系统和方法。在一个实施例中,通过配置为处理二维代码的图像数据以解码密钥信息的终端对用户进行认证,该二维代码包括用户凭证与存储在安全服务器上的数据库用户信息的加密绑定,该用户凭证包括用户面部的低分辨率图像以及可选地,用户生物特征数据。二维代码的哈希被设置为与存储在安全服务器上的用户信息的哈希具有共同的数个数位。通过计算和比较哈希值,将从安全服务器获取的用户面部的高分辨率图像与用户以及嵌入在二维代码中的低分辨率图像进行比较,以进行认证。在其它实施例中,该二维代码是在用户设备上生成的,作为交易的证明,并且可以随后用作入场券或商业合同的输出。 【EN】Systems and methods for user identity and transaction authentication are described. In one embodiment, a user is authenticated by a terminal configured to process image data of a two-dimensional code to decode key information, the two-dimensional code including an encrypted binding of a user credential including a low resolution image of a user's face and, optionally, user biometric data, to database user information stored on a secure server. The hash of the two-dimensional code is set to have a number of digits in common with the hash of the user information stored on the secure server. By calculating and comparing the hash value, the high-resolution image of the face of the user acquired from the security server is compared with the user and the low-resolution image embedded in the two-dimensional code for authentication. In other embodiments, the two-dimensional code is generated on the user device as proof of the transaction and may be subsequently used as an output for a ticket or commercial contract.
详细信息 下载全文

申请号:201911279317.2 公开号:CN111009177A 主分类号:G09B25/02
申请人:【中文】同济大学【EN】TONGJI University 申请日:2019.12.13 公开日:2020.04.14
摘要:【中文】根据本发明的用于骨钻孔教学的三坐标式机器人,包括三坐标机械臂部、末端执行部、示教器、控制部。三坐标机械臂部包括均具有导轨、模组滑块的X轴模组、Y轴模组、Z轴模组,X轴模组具有水平设置的X轴导轨、X轴模组滑块,Z轴模组具有竖直设置的Z轴导轨、Z轴模组滑块,Z轴导轨的下端与X轴模组滑块相连,X轴模组滑块带动Z轴模组沿X轴导轨移动,Y轴模组具有垂直于X轴和Z轴平面设置的Y轴导轨、Y轴模组滑块,Y轴模组滑块与Z轴模组滑块相连,Z轴模组滑块带动Y轴模组沿Z轴导轨移动,控制部控制第一摄像头、第二摄像头对模拟骨进行定位,并控制电钻对模拟骨进行钻孔后,将力传感器得到的数据,同步显示在人机互动界面上。 【EN】The three-coordinate robot for bone drilling teaching comprises a three-coordinate mechanical arm part, a tail end execution part, a demonstrator and a control part. The three-coordinate mechanical arm part comprises X-axis modules, Y-axis modules and Z-axis modules, wherein the X-axis modules are provided with X-axis guide rails and X-axis module sliders, the X-axis modules are provided with Z-axis guide rails and Z-axis module sliders, the Z-axis guide rails are vertically arranged, the lower ends of the Z-axis guide rails are connected with the X-axis module sliders, the X-axis module sliders drive the Z-axis modules to move along the X-axis guide rails, the Y-axis modules are provided with Y-axis guide rails and Y-axis module sliders, the Y-axis module sliders are perpendicular to X-axis and Z-axis planes and connected with the Z-axis module sliders, the Z-axis module sliders drive the Y-axis modules to move along the Z-axis guide rails, the control part controls the first camera and the second camera to position the simulated bones and controls the electric drills to the simulated bones, and data obtained by the force sensors are synchronously displayed on a human-machine interaction interface.
详细信息 下载全文

申请号:201910768941.2 公开号:CN111210760A 主分类号:G09G3/32
申请人:【中文】成都芯源系统有限公司【EN】Monolithic Power Systems, Inc. 申请日:2019.08.20 公开日:2020.05.29
摘要:【中文】本公开的实施例揭露了用于LED阵列驱动系统的通信接口。该通信接口包括数据写传输结构,至少包含地址编码以识别LED阵列的多个LED中与该地址编码对应的LED、读/写标识编码用于指示该通信接口是否工作于写模式、和灰度命令编码用于表示与所述地址编码对应的LED的灰度命令值。该通信接口还包括数据读传输结构,至少包含地址编码、读/写标识编码、和灰度反馈编码用于表示与所述地址编码对应的LED的灰度命令值。该通信接口可以耦接主控制器,实现主控制器与LED阵列驱动系统之间的高效精确数据读/写传输。 【EN】Embodiments of the present disclosure disclose a communication interface for an LED array driving system. The communication interface comprises a data writing transmission structure, at least containing an address code to identify the LED corresponding to the address code in a plurality of LEDs of an LED array, a read/write identification code for indicating whether the communication interface works in a writing mode, and a gray command code for representing the gray command value of the LED corresponding to the address code. The communication interface also comprises a data reading and transmitting structure at least comprising an address code, a reading/writing identification code and a gray level feedback code for expressing the gray level command value of the LED corresponding to the address code. The communication interface can be coupled with a main controller to realize efficient and accurate data read/write transmission between the main controller and the LED array driving system.
详细信息 下载全文

申请号:201911037975.0 公开号:CN110884869A 主分类号:B65G47/52
申请人:【中文】淮北矿业股份有限公司【EN】Huaibei Mining Industry Co., Ltd. 申请日:2019.10.29 公开日:2020.03.17
摘要:【中文】本发明公开了一种煤分选运输系统,包括用于分选煤的旋流器,煤运输装置,机头漏槽和位于该机头漏槽底部的中煤仓和精煤仓,旋流器的一侧设有原煤输入口,对应的另一侧设有精煤输出口,底部设有中煤输出口,煤运输装置的一端位于精煤输出口和中煤输出口的底部,另一端位于机头漏槽的顶部,该改进结构简化了原先的运输方式,降低了事故发生的可能性和生产成本,减少岗位人员及设备维修投入,节约了电耗。 【EN】The invention discloses a coal sorting and transporting system, which comprises a cyclone for sorting coal, a coal transporting device, a machine head chute, a medium coal bin and a clean coal bin which are positioned at the bottom of the machine head chute, wherein one side of the cyclone is provided with a raw coal input port, the other corresponding side of the cyclone is provided with a clean coal output port, the bottom of the cyclone is provided with a medium coal output port, one end of the coal transporting device is positioned at the bottom of the clean coal output port and the medium coal output port, and the other end of the coal transporting device is positioned at the top of the machine head chute.
详细信息 下载全文

申请号:201911037254.X 公开号:CN110905040A 主分类号:E03B3/02
申请人:【中文】淮北矿业股份有限公司【EN】Huaibei Mining Industry Co., Ltd. 申请日:2019.10.29 公开日:2020.03.24
摘要:【中文】本发明涉及一种用于煤坑的蓄水系统,包括水井、水泵和消防管道,所述水井中设置有水泵和水泵出水管,所述消防管道中水体的流向为自水井附近流向煤坑附近,该消防管道中设置有串联连接的第一三通接件、第二三通接件和单向阀,水泵出水管末端通过第一三通接件与水井附近的消防管道连接,煤坑附近底面设置有一个压滤水仓,压滤水仓通过管道和第二三通接件与水井附近的消防管道连接,本发明的有益效果是:水井中蓄积雨水得到贮存和利用,为煤坑附近的消防、生产提供了备用水,缓解了煤坑周边选煤厂和发电厂的用水压力,有效减少了选煤厂的经济损失。 【EN】The invention relates to a water storage system for a coal pit, which comprises a water well, a water pump and a fire fighting pipeline, wherein the water well is internally provided with the water pump and a water pump water outlet pipe, the flow direction of water in the fire fighting pipeline is from the vicinity of the water well to the vicinity of the coal pit, the fire fighting pipeline is internally provided with a first tee joint piece, a second tee joint piece and a one-way valve which are connected in series, the tail end of the water pump water outlet pipe is connected with the fire fighting pipeline near the water well through the first tee joint piece, the bottom surface near the coal pit is provided with a filter-pressing water sump, and the filter-pressing water sump is connected with the fire fighting pipeline near the water well through: the rainwater accumulated in the water well is stored and utilized, so that the standby water is provided for fire control and production near the coal pit, the water pressure of a coal preparation plant and a power plant around the coal pit is relieved, and the economic loss of the coal preparation plant is effectively reduced.
详细信息 下载全文

申请号:201911356888.1 公开号:CN110983194A 主分类号:C22C38/04
申请人:【中文】燕山大学【EN】Yanshan University 申请日:2019.12.25 公开日:2020.04.10
摘要:【中文】一种超级韧性钢铁材料及其制造方法,属于钢铁材料及其加工制备领域,具体涉及一种低温条件使用的超级韧性钢铁材料及其制造方法。钢铁材料包括以下重量百分比的化学元素:0.10~0.15%C,29.5~31.5%Mn,其余为Fe和不可避免的杂质。制造方法包括以下步骤:A1、氩气保护熔炼,电渣重熔处理;A2、热轧或热锻;A3、经900℃~1100℃退火1小时,淬火;A4、冷轧,冷轧板经700℃~1200℃退火1小时,退火后进行淬火。钢材组成元素简单,不含贵金属;平均晶粒尺寸小于30微米,具有完全面心立方结构,无磁性。低温条件下性能尤其突出,低温冲击功超过目前已知所有金属材料。 【EN】A super-toughness steel material and a manufacturing method thereof belong to the field of steel materials and processing and preparation thereof, and particularly relate to a super-toughness steel material used under a low-temperature condition and a manufacturing method thereof. The steel material comprises the following chemical elements in percentage by weight: 0.10 to 0.15% of C, 29.5 to 31.5% of Mn, and the balance of Fe and inevitable impurities. The manufacturing method comprises the following steps: a1, argon protection smelting and electroslag remelting treatment; a2, hot rolling or hot forging; a3, annealing at 900-1100 ℃ for 1 hour, and quenching; a4, cold rolling, annealing the cold-rolled sheet at 700-1200 ℃ for 1 hour, and quenching after annealing. The steel has simple composition elements and does not contain noble metals; the average grain size is less than 30 microns, and the crystal has a complete face-centered cubic structure and is nonmagnetic. The performance is particularly outstanding under low-temperature conditions, and the low-temperature impact energy exceeds that of all the metal materials known at present.
详细信息 下载全文

申请号:202010005269.4 公开号:CN111182379A 主分类号:H04Q11/00
申请人:【中文】中原工学院【EN】ZHONGYUAN UNIVERSITY OF TECHNOLOGY 申请日:2020.01.03 公开日:2020.05.19
摘要:【中文】本发明提出了一种弹性光网络中可塑预留业务的频谱分配方法,对于一个可塑预留型连接请求,计算其被允许使用的频块尺寸类型;然后针对每种类型的频块尺寸,在其所在路径的允许时间范围内查找所有可用频块集合,所有尺寸类型可用频块的并集为连接请求的可用频块集合;计算每个可用频块的度量向量,度量向量由消耗资源数、影响资源数和到角落距离这三个分量组成;按照字典序比较各个可用频块的度量向量,并选择度量向量最小的可用频块分配给连接请求。本发明在为可塑预留型连接请求分配资源时,充分考虑了其所消耗的资源数量以及所分配频谱对网络剩余资源的影响,可以提高网络频谱资源的利用率、降低可塑预留型连接请求的阻塞率。 【EN】The invention provides a frequency spectrum allocation method of a plastic reserved service in an elastic optical network, which is used for calculating the size type of a frequency block allowed to be used for a plastic reserved type connection request; then, aiming at the size of each type of frequency block, searching all available frequency block sets in an allowed time range of a path where the frequency block is located, wherein the union set of all size type available frequency blocks is the available frequency block set of the connection request; calculating a measurement vector of each available frequency block, wherein the measurement vector consists of three components of consumed resource number, influence resource number and corner distance; the metric vectors of the available frequency blocks are compared according to a dictionary order, and the available frequency block with the minimum metric vector is selected to be allocated to the connection request. When the invention allocates the resource for the plastic reservation type connection request, the quantity of the consumed resource and the influence of the allocated frequency spectrum on the network residual resource are fully considered, the utilization rate of the network frequency spectrum resource can be improved, and the blocking rate of the plastic reservation type connection request can be reduced.
详细信息 下载全文

申请号:202010024509.5 公开号:CN111275244A 主分类号:G06Q10/04
摘要:【中文】本发明公开了一种车速时间序列分频预测方法,属于智能驾驶领域。首先利用经验小波变换将将原始车速时间序列分解为多个子层,然后建立EWT‑LSTM‑IEWT预测模型对最高频分量多次分解预测,并采用长短期记忆神经网络(LSTM)对频率仅低于最高频的分量进行预测,接着用核函数极限学习机对各个其余的低频分量进行预测,然后采用LSTM神经网络对最高频分量的预测误差进行预测以进一步降低最高频分量对预测结果的影响,最后对所有分量的预测结果进行逆经验小波变换(IEWT),进而得到最终的车速预测结果。本发明采用核函数极限学习机和LSTM方法分别对车速时间序列的低频和高频分量进行单步预测,能有效提高短期车速预测模型的预测精度。 【EN】The invention discloses a vehicle speed time series frequency division prediction method, and belongs to the field of intelligent driving. Firstly, decomposing an original vehicle speed time sequence into a plurality of sublayers by using empirical wavelet transformation, then establishing an EWT-LSTM-IEWT prediction model to perform multi-decomposition prediction on a highest frequency component, predicting a component with a frequency lower than the highest frequency by using a long-short term memory neural network (LSTM), predicting each other low-frequency component by using a kernel function limit learning machine, predicting a prediction error of the highest frequency component by using the LSTM neural network to further reduce the influence of the highest frequency component on a prediction result, and finally performing Inverse Experience Wavelet Transformation (IEWT) on the prediction results of all components to further obtain a final vehicle speed prediction result. The invention adopts the kernel function extreme learning machine and the LSTM method to respectively predict the low-frequency component and the high-frequency component of the vehicle speed time sequence in a single step, and can effectively improve the prediction precision of the short-term vehicle speed prediction model.
详细信息 下载全文

36 条,当前第 1-10 条 下一页 最后一页 返回搜索页