课程文献中心 更多>>

关键词： Closed-Form   抠图   线缆绝缘   前景像素   背景像素   不透明度  

摘要： 针对当前自动化设备存在的因线缆绝缘边缘检测精度不足导致厚度测量不精确的问题,提出一种基于Closed-Form抠图算法的绝缘边缘检测方法。将成像中的绝缘部分视为前景像素,垫片部分视为背景像素,对成像施加一定的约束条件,通过Closed-Form抠图算法估算未知像素的不透明度,实现绝缘像素精确提取,最终测量出线缆的绝缘厚度。仿真与试验结果表明,文中所提方法能够有效提取线缆绝缘边缘,提升绝缘厚度的测量精度。

关键词： 信息科技   计算思维   “学思并行”  

摘要： 我国教育质量的持续提升激发了社会对优秀人才的高度关注。小学教育阶段作为学生技能培养的起步阶段,对于人才的培育产生了根本性的影响。这一重要时期,教师应基于学生的成长特性,精心设计教学策略,最大限度地利用信息科技教育课程,激发学生的学习热情;“学思并行”,引导他们对所学知识进行消化、吸收和升华并应用于现实问题的解决,以此发展学生的计算思维,全面提升他们在信息科技方面的综合素质。

关键词： 算法推荐技术   高校   思想政治理论课  

摘要： 算法推荐技术与高校思想政治理论课的结合是新时代高等教育发展的必然趋势与应然举措。在思政课教学实践中,算法推荐技术的介入提升了教学对象识别精度、丰富了教学内容供给方式、强化了教学反馈评估效果,同时也带来了资源供给失衡、主流话语权威弱化、教学内容同质化、议题设置异化等现实问题。对此,亟需从信息平台搭建、文化阵地建设、媒介环境优化以及主体素养培育等层面构建应对路径,切实发挥算法推荐技术对高校思想政治理论课的赋能增权作用。

关键词： 数字五维   设备管理   数据驱动  

摘要： 设备是企业生产制造的“基石”,是提升生产效率、提高产品质量、降低制造成本的关键环节,是提升企业核心竞争力的重要手段。近年来,烟草行业提出了“以加快新一代信息技术与制造业深度融合为主线,以推进智能制造为主攻方向,大力发展智能制造,推进智能工厂建设”的设备管理指导意见。设备是企业数字化转型和智能工厂落地的关键载体,也是连接智能制造技术与生产制造各个环节的桥梁,将设备管理体系与“云大物移智”等新技术有机融合是关系到企业智能制造和数字化转型成功与否的关键。基于此,本文提出了一种基于数据驱动的“数字五维”设备管理新模式,通过构建基于数据为核心生产要素的设备智能运维管理新模式,围绕设备精益管理体系优化重塑,将“精益+数字”有机融合,赋能设备精益管理水平持续提升,促进生产制造企业设备管理数字化转型升级。

关键词： 《黄帝四经》   独   观   知   明  

摘要： 《黄帝四经》具有较丰富的黄老思想特色的“观”与“独”的表述。“独”与“观”不仅是圣王不可或缺的政治品质,亦是圣王政治实践活动的重要面向。“独”与“观”二者具有循环互证关系:“独”所开启的“明”为“观”提供源初视域;借此视域,“观”的过程亦是圣王积累政治实践知识的重要途径。“知”的蕴蓄与沉淀,可进一步助益圣王对“本体之独”即“道”的通达。在“独”与“观”的循环互证中,还包含“知”与“明”的循环互证。借助“知”与“明”的中介功能,最终落实“独”与“观”的循环互证关系。“独”与“观”、“明”与“知”的二重嵌套结构的动态循环、互相加强的过程,亦是圣王的政治治理得以不断自我更新、长久维持的内在原因。

关键词： blade asynchronous Vibration   blade tip timing (BTT)   engine order search (EOS)   parameters identification  

摘要： Blade tip timing (BIT) technology is currently the trend in condition monitoring and fault diagnosis of rotating blades in major equipments. Nonetheless, the BTT technique, with the eharacteristics of non-uniformity and under-sampling, presents challenges in identifying the blade Vibration parameters. To address the issue of asynchronous Vibration parameter identification in rotating blades, initially, the fast Fourier transform (FFT) algorithm was utilized to extract the non-integer engine order of the amplitude and frequency of blade asynchronous Vibration. Subsequently, an improved multiple signal Classification (MUSIC) algorithm was employed to propose an integer engine order search strategy for blade asynchronous Vibration frequency based on the engine order search-based multiple signal Classification (EOS-MUSIC) algorithm. Finally, an asynchronous Vibration parameter identification algorithm for rotating blades was proposed based on the EOS-MUSIC algorithm and the FFT algorithm. The MATLAB Software was utilized for simulating the signals of blade asynchronous Vibration, and the feasibility and reliability of the proposed algorithm were validated by comparing with the existing MUSIC algorithm. Experiments on impeller blade Vibration were conducted on a large-scale centrifugal compressor test rig. The absolute error of the frequency identification is 3. 36 Hz, and the relative error is only 0.53% compared with the results of the strain gauge method. Based on the preprocessing of FFT algorithm, the paper extracts the blade asynchronous Vibration parameters by engine order search, which overcomes the problems of long calculation period and severe identification errors of the existing MUSIC algorithm. The study provides a theoretical support for the asynchronous Vibration parameters identification of rotating blades. © 2025 Chinese Vibration Engineering Society. All rights reserved.

关键词： Shrinkage  

摘要： Fast iterative shrinkage threshold algorithm (FISTA) based on fast Fourier transform (FFT-FISTA) has higher computational efficiency, but it ignores spatial Variation and winding error of point spread function to cause loss of sound source recognition Performance. Here, an improved FFT-FISTA algorithm based on function beamforming was proposed. This improved algorithm could take Output of function beamforming as iterative input of FFT-FISTA to establish a linear equation System for function beamforming, sound source distribution and rising power spatial transfer invariant point spread function. The linear equation System was iteratively solved based on FFT under periodic boundary conditions to make a calculated non-periodic function become a periodic function, solve the problem of wave number leakage brought by zero-filling boundary, improve the accuracy of Operation and further enhance imaging Performance. The main lobe of point spread function was sharpened through exponential Operation, and the applicability of the assumption of spatial transfer invariance of point spread function was expanded. Simulation and test results showed that compared with the conventional FFT-FISTA, the improved FFT-FISTA can improve imaging spatial resolution and dynamic ränge, expand the effective imaging area of FFT-FISTA;the compressed gas leakage experiment results verify the effectiveness of the improved FFT-FISTA. © 2025 Chinese Vibration Engineering Society. All rights reserved.

关键词： corpuscular method   flexible cylinder ejection   Kriging Surrogate model   non-dominated sorting genetic algorithm II (NSGA-II)  

摘要： As a new type of ejection method, flexible cylinder ejection has advantages of infrared target concealment and stable energy Output. Here, to solve problems of larger overload and longer response time of flexible cylinder ejection, and further improve ejection response speed and ejection stability, a Surrogate model optimization method was proposed to optimize ejection process of flexible cylinder, reduce ejection overload and improve ejection speed. Based on the Surrogate model theory, a flexible cylinder ejection Surrogate model was established, and the model's accuracy was analyzed. Then, specific effects of 3 key parameters of diameter of inflation hole, opening time and opening duration on ejection dynamic response were deeply explored. Relevant parameters of the ejection model were optimized using the optimization algorithm called non-dominated sorting genetic algorithm II (NSGA-II). The study results showed that the finite element model using the corpuscular method can accurately simulate ejection process of flexible cylinder;however, compared to the response surface model, Kriging Surrogate model exhibits higher accuracy in replacing the finite element model of flexible cylinder;aiming at initial design points, a balanced design scheine optimized with NSGA-II was proposed by authors, this scheme can successfully increase the ejection speed by 4.79% and reduce the ejection overload by 21.70%;an optimization scheme is proposed for optimization processes of ejection speed and maximum overload. © 2025 Chinese Vibration Engineering Society. All rights reserved.

关键词： 改进RRT^(*)   自适应步长   偏向目标采样   路径拉伸   路径平滑   路径规划  

摘要： 针对机器人路径规划问题,设计了一种基于改进RRT^(*)算法的机器人路径规划方法。首先,针对RRT^(*)算法环境适应能力差的问题,提出基于奖励的自适应步长设计,以提高RRT^(*)算法的环境适应性;其次,针对RRT^(*)算法在采样空间随机采样的特性,提出偏向目标采样与贪心采样结合的采样策略,并设置路径长度阈值,对采样点进行控制;最后,针对路径冗余且存在尖角的问题,采用先对路径进行拉伸处理再进行平滑处理的方式来解决。实验结果表明,所提出的改进RRT^(*)方法相较于其他方法,效率更高、运行时间更短,在路径的平滑处理方面表现出色。

关键词： 无线传感器   WSN覆盖率   鹅优化算法   节点初始化   动态变异   准反向学习  

摘要： 针对无线传感器(WSN)随机部署时节点分布不均、覆盖率较低的问题,提出一种基于动态变异准反向学习改进鹅优化算法(GOOSE)的WSN优化策略。动态变异准反向学习是一种较为新颖的智能优化算法改进策略,将其应用在种群初始化阶段和种群迭代过程中,可以显著提高初始节点和迭代寻优过程中种群的均匀度,间接提升节点的覆盖率。先将随机初始化节点与加入动态变异准反向学习后的初始化节点进行对比,再使用两种算法的目标函数分别对8个测试函数进行仿真实验,最后对改进前后的GOOSE-WSN覆盖率图以及覆盖曲线进行对比。实验结果表明,改进后的鹅优化算法(DGOOSE)性能明显更优,在WSN覆盖优化方面也有着不错的效果,证明了研究的可行性、有效性和先进性。