摘要
为了改善传感器的动态响应特性,提出了一种动态补偿器的设计方法.该方法在传感器后接一基于模型参考自适应方法的动态补偿器,并利用BP神经网络算法不断在线辨识补偿器参数,使其始终工作在最佳参数下,实现动态性能的在线补偿.试验表明,对动态品质较差的双孔悬臂梁式压力传感器进行在线补偿,其到达稳态值的调节时间由0.1s缩短为0.015s,动态性能得以较好的改善.该方法为提高传感器性能开辟了新途径.
In order to improve dynamic response performance of sensor, a dynamic digital compensation method is presented. In model reference adaptive method. A dynamic compensated digital filter based on model reference is connected with the sensor. The BP neural networks algorithm is used to identify compensation parameters. The dynamic compensated digital filter works at state of the optimum parameter all the time, that the dynamic performance of the sensor is compensated on-line. Experimentation proves that the double-hole-cantilever girder-pressure sensor that has the bad dynamic performance is compensated on-line, the adjust time of the system is shorted from 0.1 s to 0.015 s and the dynamic performance is improved. The method adopted was useful and referential for improving the dynamic performance of sensor.
出处
《传感技术学报》
CAS
CSCD
北大核心
2007年第10期2284-2287,共4页
Chinese Journal of Sensors and Actuators
基金
20年国家民委资助项目成果资助(2004M005)
西北第二民族学院院级科研项目资助(2004Y023)
关键词
传感器
动态补偿
模型参考
BP神经网络
Sensor
Dynamic Compensation
Model Reference
BP neural networks
