3种空隙率测量建模方法的比较研究  

Comparative study on three modeling methods for voidage measurement

作　　者：彭珍瑞[1] 魏薇[1] 李国娜[2] 吉万成[2] 

机构地区：[1]兰州交通大学机电工程学院,甘肃兰州730070 [2]中国石油塔里木油田分公司,新疆塔里木841000

出　　处：《传感器与微系统》2009年第8期49-51,55,共4页Transducer and Microsystem Technologies

基　　金：甘肃省自然科学基金资助项目(3ZS062-B25-016);甘肃省教育厅科技计划资助项目(20865)

摘　　要：介绍了测量空隙率的电容层析成像(ECT)系统,分析了ECT测量空隙率的基本原理。简述了3种空隙率测量建模方法,即分别基于遗传算法—偏最小二乘(GA-PLS)算法、蚁群算法(ASA)和最小二乘支持向量机(LS-SVM)的建模方法。建模数据均来自ECT传感器获取的电容值。在此基础上,提出了以ECT传感器为信息获得手段的空隙率测量建模的统一框架。通过实验数据对3种方法进行了评价,对它们从泛化误差、最大误差、平均误差和测量时间几个角度进行了比较。结果表明:LS-SVM方法的精度最高,泛化能力最强。Electrical capacitance tomography（ECT） system for voidage measurement is described. Voidage measurement principle based on ECT is analyzed. Three modeling methods for voidage measurement, which are based on genetic algorithm-partial least square （ GA-PLS ）, ant system algorithm （ ASA ）, and least squares support vector machine（LS-SVM） respectively, are reviewed briefly. All the data under three voidage measurement methods come from the ECT sensor. A unifying model is put forward to provide a universal voidage measurement framework. The experimental data are used to evaluate the three methods in terms of generalization ability index, maximum measurement error,mean measurement error, and measurement time. The results show that the LS-SVM method has highest precision and best generalization ability.

关 键 词：空隙率测量 电容层析成像 遗传算法 蚁群算法 最小二乘支持向量机 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] O359[自动化与计算机技术—计算机科学与技术]