一种用于在线测量的电容层析成像图像重建算法

提出了一种新的、快速的电容层析成像算法。在分析极小范数解的基础之上针对ECT逆问题的特点对其进行改进,并利用奇异值分解定理证明这种改进的数值稳定作用,从而从极小范数解的角度解决了ECT逆问题数值解的唯一性和稳定性问题;在此基础上从最优化的角度推导出进一步提高重建图像质量的公式;数值实验表明这种改进是有效的,其图像重建时间近似LBP,然而成像质量比LBP、Tikhonov和Landweber迭代法要好。     

电容层析成像技术研究进展

在介绍了电容层析成像技术原理及组成部分的基础上,详细阐述了电容层析成像技术的研究现状,包括微小电容检测与数据采集系统、图像重建算法和应用研究,总结了电容层析成像技术在多模态成像、图像重建算法及成像质量评价等方面面临的挑战,并指出了未来努力的方向。     

基于非凸与不可分离正则化算法的电容层析成像图像重建

搅拌器内两相混合是化工生产中常见的现象,电容层析成像(ECT)技术主要对两相分布进行可视化重构,以达到监测的目的。受稀疏贝叶斯学习的启发,提出了一种非凸与不可分离正则化(NNR)算法重建ECT图像。在稀疏先验的基础上引入矩阵低秩特性,采用最大后验估计在潜在空间中提出一个新的优化问题,利用对偶变量将潜在空间的目标函数映射到原始空间进行迭代求解,用来恢复同时稀疏与低秩的矩阵。与凸近似L1范数相比,NNR算法可获得更准确的重建图像,同时比非凸可分离方法更容易收敛到全局最优解。为验证NNR算法的重建效果,通过数值仿真与静态实验的方法分别与其他5种算法进行重建对比。结果表明：NNR算法可以有效减少重建伪影,提升中心物体的重建质量,为搅拌器内两相分布提供了高质量的重建算法。     

基于电容层析成像多相流检测的动态重建算法研究

为了提高ECT图像重建质量,提高ECT在线监测的精确性,提出了一种广义动态图像重建算法(DR),集合了ECT测量信息和流动对象的物理演变信息。该算法同时考虑被测对象的空间约束、时间约束以及动态变化信息的目标功能。动态实验研究表明,该算法成功解决ECT逆问题,有效突破了ECT图像重建的数值不稳定性,其成像质量明显好于一般的静态图像重建算法。     

电容层析成像测量气固两相流的影响因素

气力输送管道内经常出现的磨损现象和被测介质的不透明性,使得运用常规方法测量固体流动时的参数困难重重。ECT作为最先进和最有效的方法之一,可在线测量流动管道内固体横截面浓度分布和固体速度,便可计算得到固体质量流量。然而在大部分实际应用中,存在许多因素会影响管内固体的横截面浓度分布和固体的湿度值,使计算质量流量时产生误差。本文通过探讨找到解决此问题的方法,使得计算误差减小到可以忽略,结果更贴近实际情况。     

基于卷积神经网络的医学图像分割的数据增强方法研究

基于脑损伤图像分割的临床诊断正处于快速发展阶段,利用卷积神经网络(Convolutional neural networks,CNNs)可以自动多次地完成脑损伤图像分割。数据增强可以用于改进CNNs的训练,但脑损伤图像分割的数据增强需要考虑到病变区域。我们提出了一种简单的数据增强方法,称为病变镜像,不同于基于“混合”的数据增强方法,如Mixup和CutMix。病变镜像针对病变区域,以3D图像中心为原点,对于X、Y、Z3个坐标轴而言,具有同等的1/2概率选择是否镜像,从而形成新的图像以及对应的注释。生成新的图像可用于改进CNNs的训练。为了评价所提出的方法,我们在一个脑损伤数据集上进行了实验,结果表明,在大部分情况下,与Mixup、CutMix和传统数据增强(Traditional Data Augmentation,TDA)相比,该方法提高了分割精度。     

一种新的电容层析成像电极组合激励测量模式

电容层析成像图像重建为非线性不适定反问题,其敏感场分布不均匀,可获得的测量数据有限,中心处物体的成像效果不佳。增加电极数目,可获得更多的电容测量数据,减小其不定性,同时改善敏感场分布。但同时导致电容测量值变小,测量精度下降。在保证电容测量精度的前提下,提出了24电极组合式电容层析成像传感器结构,研究了两种激励测量方案,并与传统12电极电容层析成像传感器进行了对比分析,包括电容测量值的大小及其动态测量范围、灵敏场分布的均匀性以及不同流型下的重建图像。仿真结果表明,与12电极电容层析成像传感器相比,采用24电极组合式电容层析成像传感器,其电容测量值大,可较好地保证测量精度,其灵敏度分布更加均匀,对中心处物体的成像质量明显提高。     

电容层析成像技术在线测量气固流化床空隙率的研究

基于电容层析成像技术，提出了一种在线测量气固流化床空隙率的新方法。建立了相应的12电极电容层析成像气固流化床空隙率测量系统，可同时实现气固 流化床空隙率分布的在线显示和整体空隙率测量。选择加权反投影算法进行图像重建以保证空隙率分布显示的实时性和有效性。采用Tikhonov正则化原理和ART算法相结合的组合型新图像重建算法来实现整体空隙率的测量。Tikhonov正则化原理用于克服图像重建过程中的不适定问题，ART算法用于提高最终重建图像的质量。研究表明以上提出的空隙率测量新方法是有效的。空隙率分布在线测量的速度可达25幅／秒以上，整体空隙率测量的最大误差可小于5％。     

基于卷积神经网络的轮胎X光图像缺陷检测

为解决常用轮胎X射线图像缺陷检测方法难以获取准确的图像特征的问题,提出一种通过卷积神经网络获取图像特征的方法。对轮胎X射线图像进行数据增强,然后建立网络模型。训练算法获取图像缺陷特征,并用训练好的模型识别图像中的缺陷。首先将参数对应的神经元分为关键和非关键部分,然后采用局部关键点和动态学习率实现参数快速调节。试验结果表明,设计的网络模型不易过拟合,参数调节快,所需时间短,检测准确率高。     

基于轻量化卷积神经网络的高分辨率遥感图像地物分类研究

近年来深度学习技术得到快速发展,其逐渐被应用于高分辨率遥感图像的地物分类中。深度学习方法通过自动学习遥感图像中的高层次特征进行地物信息提取,较传统方法能取得更好的分类效果。Transformer架构已逐渐成为遥感图像的信息提取领域主要的深度学习模型,具有对图像的分类精度高的优点,但其模型往往结构复杂、计算量大、推理速度慢,导致应用场景受限。本文提出了一个轻量化卷积神经网络模型方法（ConvNeXt-Tiny+Multilayer Perceptron,ConvN-T+MLP）用于遥感图像地物分类,旨在保持或提高模型分类精度的同时降低模型复杂度。实验结果表明,本文提出的Conv N-T+MLP模型在具备较低计算量和参数量的同时,提高了遥感图像的地物分类精度,体现出模型的轻量性、高效性。     

采用二阶导数阵作为正则化的电容成像图像重建算法

Tikhonov正则化方法是解决病态逆问题的常用方法,正则化项的引入能改善问题的病态性。利用Laplace算子对正则化项中所包含的图像信息进行锐化处理,可提高电容成像图像重建的质量。仿真结果表明,相对于标准形式的Tikhonov正则化方法,采用正则算子为二阶导数算子的正则化方法图像重建结果边缘及轮廓清晰,对于各种设定流型均具有良好的适应能力,且图像重建结果对初始设定解不敏感,图像重建结果质量更高。     

Measurement of particle concentration in a Wurster fluidized bed by electrical capacitance tomography sensors

It is essential to measure and monitor the particle flow characteristics in a Wurster fluidized bed to understand and optimize the coating processes. In this article, two electrical capacitance tomography (ECT) sensors are used to measure the particle concentration in different regions in a Wurster fluidized bed for the “cold” particle flows. One ECT sensor has a 12‐4 internal‐external electrodes and another has eight electrodes. The 12‐4‐electrode ECT sensor is used to measure the particle concentration in the annular fluidization region (outside of the Wurster tube) and the eight‐electrode ECT sensor is used to measure the particle flow in the central region (inside the Wurster tube). The effect of particle type, particle moisture, fluidization velocity, and geometrical parameters on the Wurster fluidization process is studied based on the two ECT measurements. The radial particle concentration profiles in the annular fluidization and central flow regions with different operation parameters are given. Fast Fourier Transform analysis of the particle concentration in the Wurster tube is performed with different superficial air velocities. The optimum operating ranges of the Wurster fluidization process for different particles are given. In the end of the article, computational fluids dynamics simulation results are given and used to compare with the measurement results by ECT for a typical Wurster fluidized bed. © 2014 American Institute of Chemical Engineers AIChE J 60: 4051–4064, 2014     

电容层析成像技术在分离器料腿浓度测量中的应用

Accurate solid concentration measurement plays a key role in the process industry. Measurements analyzed offline can be used to estimate process efficiencies, to identify problems in a flow, and to validate computational models. Online measurements can be used for active control. Electrical capacitance tomography (ECT) is a unique measuring technique with great potential in multiphase flow measurement. Experimental studies are carried out on a solid concentration measurement in a cyclone separator dipleg, using ECT. In this experiment eight electrodes are selected for the ECT sensor that is placed on the straight tube of the dipleg. The fluctuating characteristics according to the screw feeder and the effect of the airflow rate from the top of the cyclone are analyzed. The feasibility and reliability of the method are verified by the experimental results.     

旋风筒中磷石膏颗粒浓度分布的电容层析法实验研究

采用双层电容层析法在线测量旋风筒中磷石膏颗粒的径向分布浓度,考察风速、气相中磷石膏颗粒质量浓度、排料口直径对颗粒浓度分布和分离效率的影响。结果表明:磷石膏颗粒在旋风筒中呈环状分布,中心处颗粒浓度小,筒壁处颗粒浓度最大;风速增大,磷石膏径向颗粒浓度减小,分离效率呈现先增大后降低的趋势,有利于改善传热效率,不利于SO_2浓度提升;气相中磷石膏颗粒质量浓度增大,磷石膏径向颗粒浓度增大,分离效率提高,不利于改善传热效率,有利于SO_2浓度提升;不同的风速存在不同的适宜排料口直径。     

基于ECT和蚂蚁算法的油气两相流空隙率在线测量

基于电容层析成像系统（ECT）和蚂蚁算法,提出了一种油气两相流空隙率在线测量的新方法。该方法利用电容层析成像系统12电极电容传感器所获取的66个测量电容信息,首先根据电容层析成像系统所获取的流型辨识结果确定对应流型下的实际空隙率测量模型参数f和b,然后利用蚂蚁算法的信息素信息,找到当前测量状态下对空隙率起主要作用的组合电容集合和相应的权重系数,从而实现空隙率测量。与流型相对应的空隙率测量线性模型参数f和b基于先验数据通过最小二乘方法确定。油气两相流的实验结果表明,该方法对空隙率的在线测量是有效的,避免了复杂的图像重建计算,实时性能佳,测量时间小于0.08 s。在几种典型流型下,提出的空隙率测量方法与常用的快关阀方法相比最大测量偏差小于5.5%。     

电容层析成像在高压浓相煤粉气力输送中的应用

Electrical capacitance tomography (ECT) was applied to dense-phase pneumatic conveying of pulverized coal, including the visualization of gas-solid flows in a horizontal pipeline.The pressure of experimental setup was up to 4.0 MPa, the solid-gas ratio was up to 11.73 kg·kg-1, and the diameter of conveying pipeline was 10 mm.The pipeline thickness of 8-electrode ECT system was 5 mm.An improved AC-based capacitance measuring circuit was developed.Single channel capacitance measuring circuit was adopted to si...     

A process fault diagnosis method using multi-time scale dynamic feature extraction based on convolutional neural network

Unlike many other techniques used in process control, which are widely applied in practice and play significant roles, abnormal situation management (ASM) still relies heavily on human experience, not least because the problem of fault detection and diagnosis (FDD) has not been well addressed. In this paper, a process fault diagnosis method using multi-time scale dynamic feature extraction based on convolutional neural network (CNN) consisting of similarity measurement, variable ranking, and multi-time scale dynamic feature extraction is proposed. The CNN-based model containing the fixed multiple sampling (FMS) layer can extract dynamic characteristics of process data at different time scales. The benchmark Tennessee Eastman (TE) process is used to verify the performance of the proposed method.     

Reconstruction of electrical capacitance tomography images based on fast linearized alternating direction method of multipliers for two-phase flow system

Electrical capacitance tomography (ECT) has been applied to two-phase flow measurement in recent years. Image reconstruction algorithms play an important role in the successful applications of ECT. To solve the il-posed and nonlinear inverse problem of ECT image reconstruction, a new ECT image reconstruction method based on fast lin-earized alternating direction method of multipliers (FLADMM) is proposed in this paper. On the basis of theoretical analysis of compressed sensing (CS), the data acquisition of ECT is regarded as a linear measurement process of permittivity distribution signal of pipe section. A new measurement matrix is designed and L1 regularization method is used to convert ECT inverse problem to a convex relaxation problem which contains prior knowledge. A new fast alternating direction method of multipliers which contained linearized idea is employed to minimize the objective function. Simulation data and experimental results indicate that compared with other methods, the quality and speed of reconstructed images are markedly improved. Also, the dynamic experimental results in-dicate that the proposed algorithm can fulfil the real-time requirement of ECT systems in the application.