基于神经网络的过程软测量  

The Process Soft-Sensing Based on Neural Networks

作　　者：孙欣[1] 王金春[1] 何声亮 

机构地区：[1]浙江大学,杭州310027

出　　处：《自动化仪表》1996年第9期7-10,共4页Process Automation Instrumentation

摘　　要：提出基于神经网络的软测量技术的一般框架,并针对一个实际的催化裂化装置,采用多层前向网络和推广随机逼近算法对粗汽油干点的软测量进行了研究.理论分析和模拟表明,该模型可以很好地描述实际对象特性.In this paper, the general framework of the soft-sensing technology based on neural networks is proposed and the soft-sensing of the naphtha endpoint is studied by means of a multilayer feedforward and the generalized chemotaxis algorithmfor the practical fluidized catalytic cracking unit. The theoretical analysis and simulation show that the practical characteristic of objects can be described very well by the model.

关 键 词：软测量 神经网络 非线性模型 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]