第19卷第7期 2011年7月
文章编号1004-924X(2011)07-1588-08
光学精密工程
Optics and Precision Engineering
Vol.19No.7
Jul.2011
基于遗传优化小波神经网络逆模型的油水测量
张冬至1.2*，胡国清2
（1.中国石油大学（华东）信息与控制工程学院，山东青岛266555；
2.华南理工大学机械与汽车工程学院，广东广州510641)
摘要：考虑基于传统的介电常数法动态测量原油含水率时存在多变量交叉敏感性，检测精度无法满足石油生产实时优化控制的需要，研究了利用多传感技术对存在交叉耦合的多敏感参量进行测量，提出了一种基于多维数据驱动的遗传优化小波神经网络逆模型及其辨识方法。该模型克服了传统神经网络初始参数随机选取的盲目性，具有全局优化和复杂非线性自学习性能，摒弃了多影响因素之间的交叉敏感性。仿真和实验结果表明了该模型的有效性，其模型预测值与实验标定值之间的相关系数为0.9996，优于BP-NN模型。该方法具有较强的泛化能力和鲁棒性，有效抑制了温度、矿化度
等多参量交叉敏感性及传感器自身非线性对测量精度的影响，改善了多传感器系统的非线性动态特性和检测精度，关键词：小波神经网络；逆模型；模型识；遗传优化；油水测量
中图分类号：TP183;TP274
文献标识码：A
doi;10.3788/OPE.20111907,1588
Measurementofoil-waterflowbased oninversemodelof
wavelet neural network with genetic optimization
ZHANG Dong-zhil-2*, HU Guo-qing?
(1.College of Information and Control Engineering,China University of Petroleum(East China)，
Qingdao266555,China;2.School ofMechanicalandAutomotiveEngineering,
SouthChinaUniversityofTechnology，Guangzhou51064l,China）
CorrespondingauthorE-mail.dz.z@mail.scut.edu.cn
Abstract: As the traditional measuring method based on dielectric coefficients shows cross-sensitivity for multi-parameters in the measurement of oil/water two-phase flows, it can not meet the require ments of real-time optimization control for petroleum production.Therefore, this paper investigates a method to measure multi-parameters with cross-sensitivity by using multi-sensing technology. It pres-ents an inverse model of wavelet neural network with genetic optimization and also researches its iden tification method. The model overcomes the blindness of initialization weight-value choice in tradition al neural networks, provides the abilities of global optimization and nonlinear self-learning, and elimi-nates the cross-sensitivity of multi-factors. The simulation and experimental results demonstrate the validity and effectiveness of the proposed model and show that the correlation coefficient between the predicted values and calibration values is O. 999 6, which is better than that of BP-NN model. The
收稿日期：2010-07-01；修订日期：2010-11-26.
基金项目：华南理工大学优秀博士学位论文创新基金资助项目（No.200903023）