Damage Detection Methods for Offshore Platforms Based on Wavelet Packet Transform

The wavelet packet transform is used for the damage detection of offshore platforms. When some damage occurs, the dynamic response parameters of the structure will shift subtly. However, in some cases, the dynamic parameters, such as dynamic response, are not sensitive, and it is very difficult to predict the existence of damage. The present paper aims to describe how to find small damage by the use of wavelet packet transform. As the wavelet packet transform can be used to quickly find the singularity of the response signal on different scales, the acceleration signal of a damaged offshore platform in the time domain is transformed through the wavelet packet. Experimental results show that the Daubechies 4 wavelet transform can be used to detect damage.

Experiment Verification of Damage Detection for Offshore Platforms by Neural Networks

In the present work, damage detection for offshore platforms is divided into three steps. Firstly, the located direction of the damaged member is detemfined by the pmbabilistic neural network with input of the change rate of normalized medal frequency. Secondly, the profile and layer of the damaged member is also determined by the pmbabilistic neural network with input of the normalized damage-signal index. Finally, the damage extent is determined by the back propagation neural networks with input of the squared change rate of modal frequency. So the size of the network and the training time can be reduced greatly. All these networks are trained with simulated data obtained from the finite element model of an experiment model. Then these trained neural networks are examined with data obtained from impulse tests on the experiment model. The experiment results show that the trained neural networks are able to detect the damaged member with reasonable accuracy.

An Efficient Adaptive Failure Detection Mechanism for Cloud Platform Based on Volterra Series

Failure detection module is one of important components in fault-tolerant distributed systems,especially cloud platform.However,to achieve fast and accurate detection of failure becomes more and more difficult especially when network and other resources' status keep changing.This study presented an efficient adaptive failure detection mechanism based on volterra series,which can use a small amount of data for predicting.The mechanism uses a volterra filter for time series prediction and a decision tree for decision making.Major contributions are applying volterra filter in cloud failure prediction,and introducing a user factor for different QoS requirements in different modules and levels of IaaS.Detailed implementation is proposed,and an evaluation is performed in Beijing and Guangzhou experiment environment.

Small tracking error correction for moving targets of intelligent electro-optical detection systems

Small tracking error correction for electro-optical systems is essential to improve the tracking precision of future mechanical and defense technology.Aerial threats,such as“low,slow,and small(LSS)”moving targets,pose increasing challenges to society.The core goal of this work is to address the issues,such as small tracking error correction and aiming control,of electro-optical detection systems by using mechatronics drive modeling,composite velocity–image stability control,and improved interpolation filter design.A tracking controller delay prediction method for moving targets is proposed based on the Euler transformation model of a two-axis,two-gimbal cantilever beam coaxial configuration.Small tracking error formation is analyzed in detail to reveal the scientific mechanism of composite control between the tracking controller’s feedback and the motor’s velocity–stability loop.An improved segmental interpolation filtering algorithm is established by combining line of sight(LOS)position correction and multivariable typical tracking fault diagnosis.Then,a platform with 2 degrees of freedom is used to test the system.An LSS moving target shooting object with a tracking distance of S=100 m,target board area of A=1 m^(2),and target linear velocity of v=5 m/s is simulated.Results show that the optimal method’s distribution probability of the tracking error in a circle with a radius of 1 mrad is 66.7%,and that of the traditional method is 41.6%.Compared with the LOS shooting accuracy of the traditional method,the LOS shooting accuracy of the optimized method is improved by 37.6%.

Crack detection using a frequency response function in offshore platforms

Structural cracks can change the frequency response function (FRF) of an offshore platform. Thus, FRF shifts can be used to detect cracks. When a crack at a specific location and magnitude occurs in an offshore structure, changes in the FRF can be measured. In this way, shifts in FRF can be used to detect cracks. An experimental model was constructed to verify the FRF method. The relationship between FRF and cracks was found to be non-linear. The effect of multiple cracks on FRF was analyzed, and the shift due to multiple cracks was found to be much more than the summation of FRF shifts due to each of the cracks. Then the effects of noise and changes in the mass of the jacket on FRF were evaluated. The results show that significant damage to a beam can be detected by dramatic changes in the FRF, even when 10% random noise exists. FRF can also be used to approximately locate the breakage, but it can neither be efficiently used to predict the location of breakage nor the existence of small hairline cracks. The FRF shift caused by a 7% mass change is much less than the FRF shift caused by the breakage of any beam, but is larger than that caused by any early cracks.

Detecting damage to offshore platform structures using the time-domain data

A new method that uses time-domain response data under random loading is proposed for detecting damage to the structural elements of offshore platforms. In our study, a time series model with a fitting order was first constructed using the time-domain of noise data. A sensitivity matrix consisting of the first differential of the autoregressive coefficients of the time series models with respect to the stiffness of structural elements was then obtained based on time-domain response data. Locations and severity of damage may then be estimated by solving the damage vector whose components express the degrees of damage to the structural elements. A unique aspect of this detection method is that it requires acceleration history data from only one or a few sensors. This makes it feasible for a limited array of sensors to obtain sufficient data. The efficiency and reliability of the proposed method was demonstrated by applying it to a simplified offshore platform with damage to one element. Numerical simulations show that the use of a few sensors’ acceleration history data, when compared with recorded levels of noise, is capable of detecting damage efficiently. An increase in the number of sensors helps improve the diagnosis success rate.

Damage Detection of Offshore Jacket Structures Using Frequency Domain Selective Measurements

损坏察觉技术为的发展近海夹克衫结构是重要的阻止灾难的事件。这份报纸使用频率回答为监视的结构的健康的目的基于的方法。在完成这项任务的努力，最小的等级不安理论的概念被利用了。现在的文章介绍有希望的方法论有效地选择频率点。达到这个目标，每个成员的形式的紧张精力比率在结构的不同自然频率被评估以便为损坏察觉识别敏感频率领域。建议方法论开创大得多的察觉效率的可能性。另外，建议方法的表演在与多重损坏的关系被评估。上述的点用二个维的夹克衫平台的数字学习被说明，并且结果证明了令人满意利用建议方法论。

Corpus of Carbonate Platforms with Lexical Annotations for Named Entity Recognition

An obviously challenging problem in named entity recognition is the construction of the kind data set of entities.Although some research has been conducted on entity database construction,the majority of them are directed at Wikipedia or the minority at structured entities such as people,locations and organizational nouns in the news.This paper focuses on the identification of scientific entities in carbonate platforms in English literature,using the example of carbonate platforms in sedimentology.Firstly,based on the fact that the reasons for writing literature in key disciplines are likely to be provided by multidisciplinary experts,this paper designs a literature content extraction method that allows dealing with complex text structures.Secondly,based on the literature extraction content,we formalize the entity extraction task(lexicon and lexical-based entity extraction)for entity extraction.Furthermore,for testing the accuracy of entity extraction,three currently popular recognition methods are chosen to perform entity detection in this paper.Experiments show that the entity data set provided by the lexicon and lexical-based entity extraction method is of significant assistance for the named entity recognition task.This study presents a pilot study of entity extraction,which involves the use of a complex structure and specialized literature on carbonate platforms in English.

Application of improved back-propagation algorithms in classification and detection of scars defects on rails surfaces

An experimental platform with bracket structures,cables,parallel computer and imaging system is designed for defects detecting on steel rails. Meanwhile,an improved gradient descent algorithm based on a self-adaptive learning rate and a fixed momentum factor is developed to train back-propagation neural network for accurate and efficient defects classifications. Detection results of rolling scar defects show that such detection system can achieve accurate positioning to defects edges for its improved noise suppression. More precise characteristic parameters of defects can also be extracted.Furthermore,defects classification is adopted to remedy the limitations of low convergence rate and local minimum. It can also attain the optimal training precision of 0. 00926 with the least 96 iterations. Finally,an enhanced identification rate of 95% has been confirmed for defects by using the detection system. It will also be positive in producing high-quality steel rails and guaranteeing the national transport safety.

基于放电电压平台研究的蓄电池寿命状态评估

为解决传统动车组镍镉蓄电池的返修方法导致部分蓄电池在触发返修条件前已性能劣化,同时大量已达返修标准的蓄电池性能并未过度衰退的问题,设计单体镍镉蓄电池全寿命加速老化实验并获取相关实验数据。首先,采用集成经验模态方法建立单体电池全寿命健康状态类别划分模型,然后运用离散小波变换消除放电电压平台数据的奇异值,进而利用极限学习机算法预测蓄电池寿命状态,最终实现对蓄电池全生命周期寿命的准确预测与健康状态评估功能。实验结果表明:相较于传统的蓄电池寿命阈值分类方法,运用集成经验模态建立的健康状态类别划分模型能有效避免蓄电池寿命末端出现误警情况。作为融合算法模型输入的放电电压平台数据易获取,基于离散小波变换的数据预处理方法可提升算法准确率近3%,最终可达到96%~98%。此外,相对于传统的神经网络模型,融合算法模型不涉及迭代,因而能兼顾算法的预测精度与计算效能。蓄电池识别健康状态的F1值为0.976 3,识别老化阶段的F1值为0.950 9,识别故障阶段的F1值为0.939 394。相较于传统的依据动车组运营里程和使用年限进而决定蓄电池是否返修的方法,融合算法模型提供了显著的评判标准,能判别蓄电池是否应该返修,并有效地识别蓄电池的健康状态,降低了动车组的运营成本,保障动车组运营安全,为电池寿命评判和检修策略的优化提供参考。

变电站多尺度异常入侵目标轻量化检测方法

智能电网的建设思路决定了远郊变电站异常入侵监测的无人化趋势,促进了变电站异常入侵智能检测方法的快速发展。现阶段尚未拥有该场景下异常入侵目标数据集,且现有的目标检测方法也未针对变电站边缘计算端进行轻量化优化设计,不适用于需要全天候实时监测的变电站边缘设备。针对上述问题,从实际应用需求出发,构建变电站异常入侵目标数据集(Dataset for Anomaly Invasion Targets in Substations,SAITD),基于YOLOv5s模型提出适用于变电站边缘检测设备的轻量化异常入侵目标检测网络YOLOv5-Substation。添加微小尺度目标特征提取层与上采样轻量化算子CARAFE,在扩大感受野的同时,充分保留特征图中多尺度目标的语义信息,从架构端提高原有模型的检测精度。基于知识蒸馏模型,使用网络剪枝(Network-slimming)策略对原有模型进行轻量化改进,在保证原模型检测精度的同时,加速模型推理。仿真实验表明,轻量化后的边缘端计算模型精度相较于YOLOv5s提高了3.3%,推理速度提升了41.9%,可为智能电网的全速运行提供强有力的数据基础、技术支撑与安全保障。

基于油液磨粒检测的机械测试技术教学探索与实践

聚焦机械测试技术教学方法与内容的探索与实践,提出一种基于油液磨粒检测的创新教学方法,设计并实现教学实验平台。以实际科研案例为基础,建立实践指导理论学习的教学模式,充分培养学生的工程思维与实践能力。利用系统化、模块化思路构建教学实验平台,实现教学与科研接轨。具体内容兼顾知识基础性与测量系统整体性,将涉及的抽象理论知识转化为直观的感性认知,促进学生对相关知识点的理解与掌握,激发学生探究具体科学问题的主动性,助力高水平复合型工程创新人才的培养。

面向轨道智能交通大学生创新训练实践教学平台设计

以轨道智能交通中的入侵异物智能检测为对象,设计了一种新型的大学生创新训练实践教学平台,提出了一种基于深度学习的轨道异物入侵检测模型,并结合人工智能和机械设计等交叉学科专业知识,自主研发了一种小型轨道异物入侵智能检测系统物理样机,实现了复杂环境下的高效轨道异物入侵检测。实验和实践结果表明,该模型很好地平衡了检测速度和精度,在NVIDIA GTX1080Ti平台上对自建轨道异物入侵数据集的平均检测精度为96.1%,检测速度为209 FPS。通过上述大创实践教学平台的设计、实施和考核,培养和激发了大学生的创新实践和团队协作能力。

新疆乌鲁木齐市米东区动物疫病检测资源共享平台人才培养项目建设

随着当前动物疫病发生特点及国家关于兽医社会化服务中第三方兽医检测实验室的推进要求,根据乌鲁木齐米东区牲畜养殖量来衡量,在米东区建设动物疫病检测资源共享平台势在必行。以建设的动物疫病检测资源共享平台为基础,与新疆农业职业技术学院共同建立产学研实验室基地,开展专业人才培养,提升服务团队成员的专业和素质水平,为乌鲁木齐米东区开展“三农”工作提供强而有力的技术保障体系,提升米东区动物疫病的整体检测能力,为当地畜禽养殖保驾护航,提高养殖效益,助理乡村振兴。

基于LabVIEW-VISA的传感器检测虚拟实验平台设计

为提升应用本科机电专业学生智能传感应用能力,设计传感器检测虚拟实验平台。提出基于LabVIEWVISA的传感器检测虚拟实验平台及实验方法,在LabVIEW-VISA应用设计基础上搭建了传感器检测虚拟实验平台整体架构;设计了传感器实验箱、MyDAQ便携式测量、万用表、应变式传感器等主要硬件;进行了数据采集模块、数据处理、输出模块及信号处理电路等软件设计,利用信号处理电路实现数据在平台中的双通道传输,实现了预期目标。结果表明,设计的传感器检测虚拟实验平台实验验证灵敏度高、线性误差小,有利于提高人才实践动手能力培养。

基于改进的YOLOv5甘蔗茎节智能识别系统

甘蔗茎节的快速准确识别对实现甘蔗田间作业和糖料产业精细化管理等任务具有重要意义,是田间产量测量的前提。为此,针对当前甘蔗人工测量成本高、人工参与量大和工作效率低等问题,设计了一种基于边缘计算的便携式甘蔗茎节智能识别系统。YOLOv5网络模型具有架构小、运行速度快等优势,通过对网络模型的进一步改进,使用K均值聚类方法调整锚框尺寸至最优,增强网络模型鲁棒性的同时也提高了茎节识别准确性;在卷积层后添加了CBAM注意力机制模块,使网络模型更加关注甘蔗茎节特征;引入VarifocalNet,使网络模型对于密集遮挡区域的甘蔗茎节有更好的识别效果。将改进后的算法移植入Jetson TX2嵌入式开发平台。试验结果表明:系统平均准确率(AP值)为89.89%,相比于之前的88.23%,提升了1.66%,单幅图像检测时间约为1.87s,视频检测速度可达15fps。系统可有效实现对甘蔗茎节的实时准确识别,为甘蔗田间产量测量提供了新的解决思路。

深海探测与驻留装备发展研究

深海探测与驻留装备是发展海洋科学的重要基础。近年来,我国自主研发的相关装备进展良好,但在国产化过程中仍面临一些“卡脖子”技术问题,亟需攻克相应装备的技术短板,推动我国“透明海洋”能力建设。本文在分析深海探测与驻留装备发展需求的基础上,围绕深海通用感知探测设备、深海运载探测通用平台、深水多功能作业平台、深海通导定位装备、深海通用作业工具、深海有人与无人设备的智能控制系统等6个重点领域,系统梳理国内外相关装备的发展现状,深入辨析我国深海探测与驻留装备发展面临的问题。本文论证提出了我国深海探测与驻留装备的9个重点发展方向,涵盖探测装备能源补给、装备系统设计及优化、高功率密度动力、原位实验研究与开发、载人装备宜居与应急救援、水下动态网络定位、水下声光多模混合通信网络、作业人机混合决策与协同控制、人机交互效能评估及优化等关键技术。研究建议,坚持科学发展、加快启动重点海域建设,坚持统筹集约、高效推进共建共用共享,坚持创新驱动、持续发展新技术新装备,以推动我国深海探测与驻留装备的高质量发展。

DEALS——追踪代币转账信息不一致

区块链使传统的交易所和借贷机构能够扩展到去中心化平台(decentralized platform,Depl),任何人都可以在没有中介机构帮助的情况下进行交易和借贷.大多数Depl都是作为运行在以太坊上的智能合约实现的,并与另一种智能合约、加密货币(即代币)交互,以实现各种功能.尽管Depl涉及价值超过350亿的美元,但人们对代币的实际转移是否如Depl预期的那样一致知之甚少.代币的实际转移与Depl期望的不一致被称为行为不一致,这种不一致的出现将导致财产的损失和用户的质疑.在这项工作中,我们迈出了调查Depl和代币之间的这种不一致的第一步.我们提出通过监控Depl和代币的核心数据结构变化,将实际的代币转移行为与Depl内部记录指示的行为进行比较,自动检测不一致.实验结果表明,在1 012 749笔交易中存在不一致行为,涉及2 871对Depl和代币,与110个Depl和2 544个代币相关.实验结果的精准度为98.0%.此外,还总结了导致不一致的10大原因,例如诈骗Depl、Depl与代币尺度不一致、锁币规则不明确等.

基于YOLO的输电线路鸟巢检测网络结构改进研究

为提高电力系统的安全运行水平,针对输电线路上的鸟巢识别问题,提出基于YOLO的输电线路鸟巢检测网络。首先通过构建GhostNet模块搭建骨干网络,并优化了特征层提取方式;随后通过改进特征金字塔连接层,并结合PANet结构构建了瓶颈网络的特征金字塔,最终搭建了YOLO-NEST网络。构建并扩充数据集进行训练,将提出的网络与其他目标检测算法进行对比,结果表明所提网络在进行输电线路的鸟巢检测时效率更高。

γ射线探测综合实验教学平台开发

针对传统γ射线探测实验装置定位精度不高、搭建麻烦、可开展的实验项目少,不利于实验重复与教学等缺点,该文设计开发了一款γ射线探测与测量综合实验教学平台,完成了平台的机械设计加工以及γ射线探测专用数字化多道分析器研制。基于该平台,可完成多个关于γ射线探测与应用的认知、基础、综合型、探究型实验项目。实践教学效果证明,该平台稳定性好,操作性强,实验结果准确,方便教师教学,能使学生更好地理解与掌握γ射线的性质、γ射线与物质相互作用的机制以及γ射线的应用。