摘要：【中文】本发明提供了一种g‑C₃N₄/CNT纳米杂化材料改性PPESK润滑薄膜及其制备方法，属于膜材料制备技术领域。本发明通过将尿素和CNT煅烧制备g‑C₃N₄/CNT纳米杂化材料改性PPESK树脂，进而制备出一种具有高性能的改性PPESK润滑薄膜。本发明制备的薄膜具有高的强度和韧性，以及突出的摩擦学性能，且反应条件温和，制备工艺简单、易操作、成本低廉，适宜于大规模生产聚合物自润滑复合薄膜。 【EN】The invention provides a g-C₃N₄A/CNT nano hybrid material modified PPESK lubricating film and a preparation method thereof belong to the technical field of film material preparation. The invention prepares g-C by calcining urea and CNT₃N₄the/CNT nano hybrid material is used for modifying the PPESK resin, so that a modified PPESK lubricating film with high performance is prepared. The film prepared by the invention has high strength and toughness, andthe obtained self-lubricating composite film has the advantages of favorable tribological performance, mild reaction condition, simple preparation process, easy operation and low cost, and is suitable for large-scale production of the polymer self-lubricating composite film.

摘要：【中文】本发明涉及一种基于FPGA的动态图处理方法，方法包括：在动态图的相邻时间戳的图镜像之间具有增量的情况下，预处理模块将后一时间戳的图镜像以任意图顶点的增量计算仅依赖于该任意图顶点的前序图顶点的方式划分为至少一个路径单元；FPGA处理模块将至少两个路径单元按照每一线程单元能够独立处理路径单元的方式存储于线程直连的片上存储器上；线程单元在将前序图顶点的状态值更新的情况下确定前序图顶点的相邻的时间戳之间的增量值，并按照路径单元确定的传递方向将增量值传递至前序图顶点相邻的后续图顶点，以对后续图顶点的状态值进行更新。 【EN】The invention relates to a dynamic graph processing method based on FPGA, comprising the following steps: under the condition that increment exists between image images of adjacent timestamps of the dynamic image, dividing the image of the later timestamp into at least one path unit in a mode that increment calculation of any vertex is only dependent on the vertex of the preamble image of the any vertex by the preprocessing module; the FPGA processing module stores at least two path units on an on-chip memory directly connected with the threads in a mode that each thread unit can independently process the path unit; and the thread unit determines an increment value between adjacent time stamps of the prologue vertex under the condition of updating the state value of the prologue vertex, and transmits the increment value to the subsequent graph vertex adjacent to the prologue vertex according to the transmission direction determined by the path unit so as to update the state value of the subsequent graph vertex.

摘要：【中文】本发明公开了一种基于电网调度控制系统的小电流接地故障选线方法，当智能电网调度控制系统监测到母线可疑接地，小电流接地故障选线辅助决策功能启动；根据不同的接地方式及消弧线圈补偿模式，将零序电流幅值法、零序电流突变法，无功突变法及IQ综合法等多个维度的选线结果进行有机融合，为单相接地故障试拉路选线提供智能化辅助决策。本发明方法不需要加装硬件设施，不受系统接地方式、消弧线圈补偿方式影响，实现单相接地故障线路的快速识别和控制，减少线路试拉次数，缩短故障处理时间。 【EN】The invention discloses a small current ground fault line selection method based on a power grid dispatching control system, wherein when the intelligent power grid dispatching control system monitors that a bus is doubtful to be grounded, a small current ground fault line selection auxiliary decision function is started; according to different grounding modes and arc suppression coil compensation modes, line selection results of multiple dimensions such as a zero-sequence current amplitude method, a zero-sequence current mutation method, a reactive mutation method, an IQ synthesis method and the like are organically fused, and an intelligent auxiliary decision is provided for single-phase grounding fault trial line selection. The method of the invention does not need to install additional hardware facilities, is not influenced by a system grounding mode and an arc suppression coil compensation mode, realizes the rapid identification and control of the single-phase grounding fault line, reduces the trial pulling times of the line and shortens the fault processing time.

摘要：【中文】本发明涉及一种彩色激光打印文件暗记识别方法，并基于此方法设计了一套暗记识别比较系统。该方法的特征在于，提取预处理后的图像中暗黄色小点的位置信息，并标定最小可重复部分为暗记，并比较暗记相似度、确定其是否属于同一类暗记。该系统包含四个模块，输入模块，图像预处理模块，暗记识别模块，暗记比较模块；输入模块读取来源于本地的图像；图像预处理模块确认最佳参数，去除干扰信息；暗记识别模块提取暗记特征；暗记比较模块对比较提取到的暗记，并给出相似度判断两个暗记是否属于同一类。目前计算机视觉和图像处理技术发展迅速，提供了实验条件，目前实验也证明提取、识别并比较纸张的暗记特征是完全可行的，达到了较高的识别精度。 【EN】The invention relates to a method for identifying a secret mark of a color laser printing file, and designs a secret mark identification comparison system based on the method. The method is characterized in that position information of small dark yellow points in the preprocessed image is extracted, the minimum repeatable part is marked as a secret mark, the similarity of the secret marks is compared, and whether the secret marks belong to the same type of secret marks is determined. The system comprises four modules, an input module, an image preprocessing module, a secret mark identification module and a secret mark comparison module; the input module reads the image from local; the image preprocessing module confirms the optimal parameters and removes interference information; the secret mark identification module extracts the secret mark characteristics; and the secret mark comparison module compares the extracted secret marks and gives out similarity to judge whether the two secret marks belong to the same class. At present, computer vision and image processing technology are developed rapidly, experimental conditions are provided, and the current experiment also proves that the extraction, identification and comparison of the secret mark characteristics of paper are completely feasible, so that higher identification precision is achieved.

摘要：【中文】本发明提供一种提高利用率的洗煤工艺，设定其中一种原煤，设定原煤重量为M时，杂质重量为m0，则精煤成分的重量为m＝M‑m0,精煤率为A＝m/M；通过传送带向洗煤机内传输预设重量的原煤M；原煤进入所述洗煤机内，下端的出煤口封闭，打开重悬浮液泵，向洗煤机内注入预设重量m1的重悬浮液，精煤及悬浮液通过过滤板进入上方，中煤及矸石下沉至底部，并通过第一油缸向下移动第一预设距离D1同时，使精煤及悬浮液充分的在上端停留，打开出煤口，使下端的液体及中煤、杂质进入下方的储煤小车，至储煤小车的重量为(m1+M)/5。 【EN】The invention provides a coal washing process for improving the utilization rate, wherein one raw coal is set, when the weight of the raw coal is set as M, the weight of impurities is M0, the weight of clean coal components is M-M0, and the clean coal rate is A-M/M; conveying raw coal M with preset weight to a coal washer through a conveying belt; raw coal enters the coal washer, a coal outlet at the lower end of the coal washer is closed, a heavy suspension liquid pump is opened, heavy suspension liquid with preset weight M1 is injected into the coal washer, clean coal and suspension liquid enter the upper part through a filter plate, middlings and gangue sink to the bottom and move downwards through a first oil cylinder for a first preset distance D1, meanwhile, the clean coal and the suspension liquid are enabled to fully stay at the upper end, the coal outlet is opened, liquid at the lower end, the middlings and impurities enter a coal storage trolley below the coal storage trolley, and the weight of the coal storage trolley is (M1+ M)/5.

摘要：【中文】本发明公开了一种接触网硬点检测系统，包括，设于轨道车辆受电弓上的受电弓特征区域，设于所述轨道车辆上的照相单元和/或燃弧获取单元，以及连接所述照相单元和/或所述燃弧获取单元的车载计算单元；其中，所述照相单元配置为获取所述受电弓特征区域的图片信息；所述燃弧获取单元配置为获取所述受电弓特征区域的燃弧特征量。本发明提供的基于机器视觉的弓网硬点检测系统获得受电弓垂向加速度，辅以燃弧特征量检测的融合弓网硬点检测方法，取得了很好的效果。 【EN】The invention discloses a contact net hard spot detection system, which comprises a pantograph characteristic area arranged on a pantograph of a rail vehicle, a photographic unit and/or an arc acquisition unit arranged on the rail vehicle, and a vehicle-mounted computing unit connected with the photographic unit and/or the arc acquisition unit; wherein the camera unit is configured to acquire picture information of the pantograph feature region; the arcing obtaining unit is configured to obtain an arcing characteristic quantity of the pantograph characteristic region. The pantograph-catenary hard point detection system based on machine vision obtains the vertical acceleration of the pantograph, and is assisted with the detection of the arcing characteristic quantity to obtain a good effect.

摘要：【中文】本发明公开一种适应高帧率的人脸追踪方法，应用于图像识别领域，针对现有的终端设备算力欠缺从而导致人脸追踪速度无法与摄像设备帧率同步的问题，本发明将人脸追踪划分为同步追踪处理和异步人脸识别两个流程，并提供了一种快速人脸识别方法，利用高耗时的异步人脸识别构建运动预测向量，从而利用运动预测向量完成同步人脸追踪，以此保证在高帧率条件下人脸追踪速度和设备帧率同步。 【EN】The invention discloses a face tracking method adaptive to a high frame rate, which is applied to the field of image recognition, and aims at solving the problem that the face tracking speed cannot be synchronous with the frame rate of a camera device due to the lack of computing power of the existing terminal device.