【作者】 冯莉；

【导师】 李岩；

【作者基本信息】 哈尔滨理工大学 ， 计算机系统结构， 2011， 硕士

【摘要】 电容层析成像技术(Electrical Capacitance Tomography,ECT)在不破坏以及不干扰被测物场的基础上,通过电容测量值重建出管道或容器内部相异介电常数的空间分布状况。它具有低成本、非侵入等优点,在多相流检测领域应用前景广阔。作为ECT系统研究的关键技术,图像重建算法的好坏直接关系着重建图像的质量和速度。本文是在基于传感器结构参数优化的基础上对图像重建算法进行了较深入的研究,主要完成了以下工作:1.深入研究了电容层析成像系统的技术特点和系统组成,从理论上分析ECT技术的工作原理,对其未来的发展做了宏观上的展望。通过分析电容层析成像系统的特点,给出了代数神经网络算法在电阻层析成像系统图像重建中的优势;2.以12电极电容层析成像系统为研究对象,分别采用Matlab和ANSYS软件编程获得各种结构参数的计算机仿真实验数据,通过对比实验数据分析各种结构参数对电容传感器性能的影响。3.对目前存在的几种典型图像重建算法进行了深入研究,针对图像重建算法的欠定性问题,提出将一种新型的神经网络图像重建算法运用到电容层析成像系统的图像重建过程中,并对该方法进行改进,将整个敏感场分布划分为六个子系统,划分后的网络降低了原始网络的规模,在算法的训练速度和成像质量特别是在流型辨识等方面有了显著的提高。4.设计了ECT图像重建仿真系统软件,利用该软件可以方便地设置圆形管道、传感器以及流型分布的各项参数,对不同参数情况下的系统环境进行快速图像重建算法的仿真研究。更多还原

【Abstract】 As a kind of non-destructive and non-intrusive measurement, electrical capacitance tomography (ECT) technique can infer the distribution of the conductivity by measuring the electric potential parameter in the sensitivity field and then get the distribution of object field in the pipe. It has a wide application prospect in two-phase flow measurement with the advantages of low cost and non-invasion etc. As the key technology of ECT system research, image reconstruction algorithm has a significant impact on quality and speed of reconstructed image. More profound study is focused on the key problems such as optimized design of transducers’structure parameter, image reconstruction algorithm. The study we have done is as follows.1. Profound study is on the technical characteristics and the system composition of ECT system. The mathematical model of capacitance sensitivity field is presented according to the principle analysis of ECT system.By macro perspective, given the future prospects of its development. By analyzing the characteristics of electrical capacitance tomography, giving the advantage of algebra neural network algorithm in the electrical resistance tomography on image reconstruction.2. Taking 12-electrodes electrical capacitance tomography systems as research objects, computer simulation data of different structure parameters is received by using Matlab and ANSYS. The affection of structure parameter on sensor performance is studied by comparing the received data, too.3. More profound study is focused on the several kinds of typical reconstruction algorithms. Proposed a new type of neural network image reconstruction algorithm applied to the process of image reconstruction in electrical capacitance tomography system, then, improve it. In view of the problem of ill-posed characteristic, we divided the whole NSSN network into six sub-systems to reduce the scale of network, improve the training speed and the image quality particularly based on the flow pattern, etc. have been significantly improved.4. We design the simulation software of ECT. On that you can easily set the parameters of circular pipe, resistance sensors and distribution of flow. It also can solve the forward problem and the image reconstruction problem.更多还原