基于嵌入式系统的软仪表及其应用

Soft Sensor Based on Embedded System and Its Application

介绍了基于嵌入式系统的软仪表的设计和开发框架,提出通用软测量软件包的数据结构,并采用双重径向基函数RBF神经网络作为软测量的核心算法。同时采用两阶段法进行神经网络训练,大大降低了软测量算法的误差。最后以催化裂化为例,介绍了如何应用该系统进行轻柴油凝固点的在线软测量。系统具有通用、开放和实时性特点,能够用于各种场合,具有较高的应用价值和推广前景。

The design and developing frame of the soft sensor based on embedded system are introduced, and the general data structure of soft sensor software package is proposed. The double RBF neural network is used as the core algorithm of soft sensing and two-phase neural network training is adopted to reduce the error of soft sensing. With the application for FCC as example, the method of on-line soft sensing of freezing point of light diesel by using this system is presented. The system features generalization, open, and real-time performance, it can be widely used in various situations for offering high applicable value and bright future.