环状两相流在线监控算法仿真研究  

An On-line Monitoring Algorithm for Annular Two Phase Flow

作　　者：马秋芳[1] 肖理庆[2] 

机构地区：[1]青岛黄海学院电信与商务学院,山东青岛266427 [2]徐州工程学院信电工程学院,江苏徐州221111

出　　处：《计算机仿真》2015年第8期439-443,共5页Computer Simulation

基　　金：国家自然科学基金资助项目(61302122)

摘　　要：电阻层析成像技术是应用于两相流领域的一种无损检测技术,它通过算法的重建图像实现对两相流的可视化测量。环状流是两相流的一种典型流型,目前其重建图像存在分辨率较低的缺点。为了提高电阻层析成像环状两相流图像重建质量,提出一种基于Landweber预迭代算法与改进遗传算法的组合图像重建算法。新算法结合区间算法与Landweber预迭代算法的重建结果生成算法初始种群,并在算法变异操作中引入改进粒子群算法。仿真结果表明,在相同实验条件下,相比线性反投影算法、修正牛顿-拉夫逊算法、Landweber预迭代算法,新算法不仅有效提高了环状两相流重建图像分辨率,而且实现了绝对图像重建,可应用于环状两相流在线监控。Electrical resistance tomography is one kind of nondestructive technique applied in the two - phase flow field, it can achieve the visible measurement by the result of image reconstruction algorithm, and annular flow is one typical flow, but the space resolution is not satisfied at present. Aiming to enhance the imaging quality of electrical resistance tomography for annular flow, a hybrid image reconstruction algorithm is proposed based on Landweber pre -iteration algorithm and improved genetic algorithm. The novel algorithm generates the initial population according to the interval algorithm and the image result of Landweber pre - iteration algorithm, and applies improved particle swarm optimization to the mutation. The simulation and experimental results demonstrate that, comparing to the linear back projection algorithm, modified Newton- Raphson method and Landweber pre- iteration algorithm, under the same experimental conditions, the new algorithm can not only improve the space resolution effectively, but also real- ize absolute image reconstruction, which can be applied to on - line monitoring for annular two phase flow.

关 键 词：电阻层析成像 环状流 遗传算法 图像重建算法 区间算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]