论文部分内容阅读
软测量技术是通过数学模型来估计工程上难以检测的变量值。由于神经网络方法能够描述高度非线性的输入输出关系，因此，基于神经网络的软测量技术已经成为很有吸引力的研究领域，它将辅助变量作为神经网络的输入，将主导变量作为其输出，通过训练网络来实现主导变量在线估计。对基于神经网络的软测量技术进行了综述并详细介绍了神经网络软仪表的结构和方法，给出了神经网络软仪表的系统开发框架，讨论了它在过程控制中的应用，对其发展作了简要的展望。
Soft-sensing technology uses mathematical models to estimate difficult-to-detect variables in engineering. As neural network method can describe the highly nonlinear input-output relationship, the soft-sensing technology based on neural network has become an attractive research field. It takes the auxiliary variable as the input of neural network and the dominant variable as its output, Through the training network to achieve the dominant variable online estimation. The soft-sensing technique based on neural network is reviewed and the structure and method of the neural network soft instrument are introduced in detail. The system development framework of the neural network soft instrument is given. Its application in process control is discussed. A brief look.