Multisensor data fusion in dimensional metrology

Multisensor data fusion in dimensional metrology is used in order to get holistic, more accurate and reliable information about a workpiece based on several or multiple measurement values from one or more sensors. The theoretical background originates in classical mathematics and statistics, in methods of artificial intelligence (AI) and in the Bayesian fusion approach. Sensor technologies and sensor characteristics influence the data fusion process and determine the gain of information compared to the application of a single sensor. Homogeneous and inhomogeneous sensor configurations lead to complementary, competitive and cooperative information integration with specific advantages depending on the application. The scope includes image fusion, tactile and optical coordinate metrology, coherent and incoherent optical measuring techniques, computed tomography as well as scanning probe microscopes.     

一种机械故障监测的信息融合

在叶片振动故障监测过程中，单传感器的间断相位法只能测得叶片振幅值信息，本文应用两个传感器的信息融合技术，通过多源信息的综合、分析和推理，得出了叶片振动频率信息，提高了传感器系统的有效性，消除了单个传感器的局限性。     

基于非线性多传感器复合定位融合算法应用研究

针对移动机器人单传感器数据短时丢失、定位精度低、传感器频率异步等问题,采用激光雷达、IMU、轮式里程计获取定位信息,提出基于扩展卡尔曼滤波和互补融合的组合数据融合方法。先通过S-G滤波算法对初始定位数据进行预处理,利用扩展卡尔曼滤波融合算法实现IMU和轮式里程计传感器的定位数据融合,得到融合数据1;再利用互补融合算法将融合数据1和激光雷达进行融合得到融合定位数据2。其中融合数据1对激光雷达进行实时补正,解决频率异步的位移偏差,从而明显提高定位精度。最后采用Gazebo仿真平台,搭建移动机器人模型以及设置传感器的基本参数,验证算法的有效性和稳定性。实验结果表明：数据融合算法提高了非线性传感器的定位精度和稳定性,并且平均定位误差在8 cm内。     

��������еĶഫ������Ϣ�ںϼ���

 多传感器信息融合是一个新的研究领域，它应用到故障诊断领域还是近几年的事。文章主要论述了多传感器信息融合技术的基本概念、工作原理及常用方法，同时还介绍了其在故障诊断中的应用实例。     

复合传感技术在激光焊接过程质量监测中的应用

激光焊接过程中出现等离子体、羽辉、声音、辐射光、电等信号,通过对这些信号进行研究并与焊接过程的稳定性及焊接缺陷进行关联,可以实现焊接质量的实时监测。对不同监测手段进行对比分析,复合传感器监测技术展现出监测信息丰富,焊接状态识别精确,焊接缺陷预测准确的优势。根据焊接过程质量监测系统复合的传感器种类,将复合传感技术分为多视觉传感器复合技术、多光学传感器复合技术、光学及视觉传感器复合技术、其他复合技术4类,综述了不同传感器复合技术的研究现状及特点,指出光学及视觉传感器复合技术对焊接状态变化敏感、采样速度快、采集信息丰富的优势。介绍了几种可以应用于激光焊接生产线的先进复合传感器监测产品,明晰了复合传感技术应用于激光焊接过程质量监测的问题及未来发展方向。创新点： (1)对不同复合传感技术的特点进行总结及对比分析。(2)对商用复合传感器监测产品的原理进行分析及应用实例介绍。     

多传感器融合技术在人民币清分机中的应用

介绍了多传感器融合技术在人民币清分机系统中的应用，重点分析了像素级和特征级数据融合过程，实验表明通过数据融合能有效地实现币值识别模块和正反面朝向识别模块的功能。     

Cooperative data fusion of transmission and surface scan for improving limited-angle computed tomography reconstruction

Limited-angle computed tomography allows faster inspection during production, but the reconstruction from limited-angle transmission data is an underdetermined problem which cannot be solved without any prior knowledge of the sample. In this paper, surface data from an optical scan is selected as prior information due to its high accuracy and availability. To incorporate this information, we have developed a new cooperative data fusion model in the compressed sensing framework. The model has been applied to numerical and experimental data and solved with a tailored algorithm. We demonstrate the benefit of the data fusion model and prove the robustness of the algorithm. The results from this study indicate that the data fusion process combines features resolved by both modalities and gives a significant increase in image quality. These improvements enable metrological measurements that are impossible with data acquired with any single modality.     

飞机机电系统效能评估方法体系研究

为实现面向飞机能量的多电飞机机电系统的稳健效能评估,从提高效能结果稳定性出发,确定了一套系统的效能评估方法体系.针对传统的效能评估活动,提出相应的改进措施来缩减每个评估节点存在的风险.其中,根据机电系统的功能结构以及飞行任务,提出了基于系统级、飞行任务级和保障级的具有三维立体多层次特征的效能评估指标体系构建模式,并依据...     

多传感器信息的决策融合法及其在电磁检测中的应用

介绍多传感器信息的决策融合原理及其常用方法，重点分析了基于目标检测理论的决策融合法和基于Dempster-Shafer(D-S)证据理论的决策融合法，并分别以融合实例讨论了它们在漏磁和涡流无损检测中的应用。多传感器信息的决策融合法有效提高了多源信息的利用程度，具有较高的可靠性和稳健性。     

基于多源数据融合的半导体晶片CMP抛光材料去除率预测

目的 对半导体晶片抛光过程中的工艺参数、耗材使用量、抛光垫状态参数等多源数据预处理后进行数据融合,建立材料去除率(MRR)预测模型,为实现半导体晶片抛光加工工艺的决策和处理奠定基础。方法 研究晶片抛光加工中的数据特点及数据融合需求,提取数据集中每个晶片加工过程中的统计特征并生成新数据集,同时引入邻域特征以应对晶片加工过程中动态因素对材料去除率的影响。提出基于深度自动编码器的多源数据融合及材料去除率预测方法。设计深度自动编码器参数,优化深度自动编码器的损失函数从而增强深度自动编码器对强相关性特征变量的重建。基于深度自动编码器进行多源传感器信号融合,降低数据维度。使用超参数搜索算法优化BP神经网络超参数,利用BP神经网络方法将融合后的数据进行半导体晶片抛光过程中的材料去除率预测。结果 采用PHM2016数据集对模型进行验证,均方误差MSE达到7.862,相关性R²达到91.2%。结论 基于多源数据的融合模型能有效预测MRR,可以对半导体晶片CMP工艺过程的智能决策与控制起到良好的辅助作用。     

基于弧压反馈的薄板GTAW智能熔透控制

以轨道交通高速高铁司机室铝合金外板为载体,围绕智能焊接关键技术,针对焊接缺陷在线监测问题开展研究. 借助工艺试验平台与焊接工艺卡开展焊接缺陷试验设计、批量数据采集、专家经验标定、数据库构建,采用卷积神经网络算法对不同类型数据构建多维信息融合模型,并对融合模型进行参数优化处理,最终完成融合模型的训练、验证和测试. 结果表明,训练后的融合模型比单一信息模型对焊接缺陷具有较好的识别结果,训练集和测试集的焊接缺陷监测精度分别为99.0%和88.3%,此监测系统的数据采集和模型响应总时间小于100 ms,能够满足工程化应用需求,提高机器人焊接的智能化水平,推动企业数字化转型升级.

     

基于多传感器信息融合的输电线路巡检机器人自主越障方法研究

针对550 kV架空输电线路巡检机器人在运行过程中对线上金具识别不稳定及自主越障效率低等问题,基于多传感器信息融合的方法,结合机器人在线越障过程的特点,提出了一种基于多传感器信息融合自主越障方法与策略,并进行了在线自主越障实验。结果表明:以巡线机器人为平台,基于多传感器信息融合的机器人自主越障策略是可行的,提高了机器人越障运动的准确性和灵活性,保证机器人安全可靠的运行。     

机器人点焊多信息融合及控制

点焊质量一直是国内外焊接国界学者致力研究的重要课题之一。目前尚未得到圆满的解决。点焊过程的复杂性、熔核形成过程的不确定性,尤其是点焊过程采用的质量监控方法的单一性决定了点焊质量控制是一项非常困难的任务。本文分析了基于单一传感器信息资源来描述焊点熔核形成过程存在的局限性,首次提出了采用多传感器信息融合控制点焊质量的思想。通过参数选择、特征信息提取,建立了点焊过程多信息融合模型。采用结合法、平均法和决     

Fatigue crack fusion in thin-sheet aluminum alloys AA7075-T6 using low-speed fiber laser welding

Reinforcing cracked aluminum structures with composite patches have been recognized as an efficient and economical method to extend the service life of cracked aluminum components. To further enhance the effectiveness of composite patches, it is envisioned that the crack can be first fused by laser welding to remove the high stress concentration at the crack front before applying the composite patch. In this paper, the feasibility of the envisioned fusion repair is investigated. A systematic approach for the fusion process design is proposed to overcome challenges related to alloy strength recovery, crack tracing, focusing position, welding speed, plate flatness, shielding gas pressure, thin-sheet factors, and skewed cracks. A thick-sheet, partial penetration model is first used to determine the starting point of laser welding conditions. A systematic approach to transfer the thick-sheet condition to successful thin-sheet welding is then presented. Based on successfully fused crack samples of AA 7075-T6, the ultimate tensile strength tests show that in average 74% of the alloy's original strength was recovered for a single-pass repair and 68% for a double-pass repair and the results are highly repeatable. It should be clear to see the benefit of the crack fusion because without crack fusion, the composite patch is bonded to a part with zero UTS at the crack region and with a high stress intensity factor at the crack front.     

Multi sensor data fusion approach for automatic honeycomb detection in concrete

We present a systematic approach for fusion of multi-sensory nondestructive testing data. Our data set consists of impact-echo, ultrasonic pulse echo and ground penetrating radar data collected on a large-scale concrete specimen with built-in honeycombing defects. From each data set, the most significant signatures of honeycombs were extracted in the form of features. We applied two simple data fusion algorithms to the data: Dempster’s rule of combination and the Hadamard product. The performance of the fusion rules versus the single-sensor testing was evaluated. The fusion rules exhibit a slight improvement of false alarm rate over the best single sensor.     

信息融合在焊缝跟踪模糊控制器中的应用

在焊接过程中,焊缝自动跟踪的精度直接影响着焊接的质量.以埋弧焊控制系统为研究对象,研究了焊缝跟踪模糊控制系统,且在该系统中,采用多个传感器同时对焊缝进行跟踪检测,并利用融合算法对其进行融合,将融合后的结暴作为模糊控制器的输入.计算机仿真结暴表明,在焊缝跟踪中,采用多传感器信息融合是合理的、可行的;且可以减少焊接过程中由传感器引起的误差对跟踪精度的影响,提高控制精度.     

基于信息融合的滚动轴承寿命状态识别研究

针对滚动轴承寿命状态识别过程中,单一传感器蕴含的信息不能全面反映寿命状态的问题,文章提出了一种基于信息融合的滚动轴承寿命状态识别方法。该方法首先采用多路卷积层提取不同传感器的数据特征信息,克服单一信息源的局限性;然后采用多层卷积、池化交替级联的方式,实现多源信息的特征值深度融合,最后采用全连接和多分类函数,实现动轴承的寿命状态识别。通过不同方法的对比实验,结果表明了所提方法能够提高滚动轴承寿命状态识别率,具有较好的可行性。     

神经网络信息融合技术在岩石挖掘机系统故障诊断中的应用

将信息融合技术与神经网络结合起来，充分利用检测到的各种故障征兆信息准确诊断岩石挖掘机系统的故障。采用混合式数据融合方法，将数据级、特征级和决策级融合通过神经网络的方法综合在一起，解决输入信息不对称性问题，使得小数据量和大数据量的信息融合成为可能。     

NDT data fusion at pixel level

No single non-destructive testing (NDT) technique or method is satisfactory to fully assess the structural integrity of a material. Indeed, each method presents some limitations in terms of defect detection and characterisation. Additionally, poor signal-to-noise ratio may make signal interpretation complex or unreliable. For these reasons, the concept of NDT data fusion — based on the synergistic use of information from multiple sources in order to facilitate signal interpretation and increase defect detection and characterisation — is expanding rapidly. A carbon fibre reinforced composite material previously subject to low energy impact damage has been non-destructively inspected using eddy current and infrared thermography. Following these non-destructive examinations, the information gathered from these two NDT methods has been fused at pixel level. Several well established and recognised data fusion processes based on statistical and probabilistic algorithms were used to combine eddy current and infrared thermographic data. The results obtained with these processes are presented in this paper and their efficiency discussed. It is shown that, in certain cases, NDT data fusion at pixel level may be adequate to increase knowledge about defect location and characterisation and to reduce ambiguity.     

数据融合在电梯群控系统中的应用

将数据融合技术引入电梯群控系统中,建立了基于数据融合技术的模糊联结聚合神经网络模型,对影响电梯调度的各因素进行充分有效的数据处理后,得到的电梯控制参数作为电梯调度单元的输入.应用实例表明了该方法是方便且有效的.