基于C++与MATLAB的橡胶门尼粘度混合智能软测量系统  

Mixed Intelligence Soft Sensor System for Rubber Mooney Viscosity Based on C++ and MATLAB

作　　者：刘美[1,2] 黄道平[2] 徐小玲[1] 

机构地区：[1]茂名学院计算机与电子信息学院,广东茂名525000 [2]华南理工大学自动化科学与工程学院,广东广州516400

出　　处：《计算机测量与控制》2010年第2期299-302,共4页Computer Measurement &Control

基　　金：国家自然科学基金项目(60704012);广东省科技计划项目(2004B10301035);茂名市科技项目(201095)

摘　　要：门尼粘度是合成橡胶生产的主要质量指标,如何在线监测门尼粘度,并实现质量的自动监控是橡胶生产工业亟待解决的问题;针对强非线性、时变和间歇的合成橡胶复杂系统,结合工艺机理分析,建立了门尼粘度的基于机理、小波神经网络和最小二乘支持向量机等多种建模方法的混合智能软测量模型,解决了门尼粘度的实时在线测量问题,并开发出基于Borland C++与MATLAB的门尼粘度软测量应用软件。乙烯厂试用结果:混合智能软测量模型预测精度高,泛化能力强,运行速度快;应用软件结构明晰,功能扩充灵活,操作界面友好,操作维护方便;该系统能稳定生产、指导操作、提高产品的优级品率,为合成橡胶生产装置的优化控制创造了条件。Mooney-viscosity is the main quality index of Synthetic Rubber. Monitoring the Mooney viscosity effectively and realizing automatic quality control of production process is an urgent problem in the rubber industry. For strong non linear, time varying and complex intermittent synthetic rubber system, we establish Mooney viscosity mixed intelligent model based on mechanism, wavelet neural network and least squares support vector machines and so on combined with the process of technical mechanism, to solve the real time online Mooney viscosity measurement and develop the Mooney viscosity soft sensor applications based on the Borland C＋＋ and MATLAB. Ethylene Plant trial Results： The mixed intelligent soft-sensor model has high predictive accuracy, strong generalization ability and fast running speed; This application software has clear structure, flexible functional expansion, friendly operating interface and easy operating maintenance; The system can. guide the operation, improve excellent products probability and create the conditions for synthetic rubber production facilities opti mal control.

关 键 词：门尼粘度 混合智能模型 软测量 BORLAND C++ MATLAB 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]