摘要
为了消除压力传感器受非目标参量的影响而呈现的非线性特性,利用小波神经网络来完成压力传感器的非线性校正.利用遗传算法对小波神经网络权阈值优化,以提高网络精确度和训练速度,设计了遗传优化小波神经网络,将该网络用于压力传感器的非线性校正.仿真结果表明该方法能有效消除非目标参量对传感器输出结果的影响.压力传感器的精度和准确度都得到提高.该系统不但可以用于各类传感器的非线性校正,还可用于其它类似系统.且设计、实现简单,适于工程应用,具有实际应用价值.
In order to emendate the nonlinear characteristic of the pressure sensor caused by the impact of non-object parameters. The wavelet neural network was used in the nonlinear emendation. Genetic algo- rithm was introduced to optimize the parameters , and the genetic wavelet neural network was put forward. The higher accuracy and faster speed was obtained. The simulation of pressure sensor shows that this system successfully eliminated the impact of non-object parameters and reflect the plant accurately and entirely. The precision and veracity of pressure sensor was increased. The system is also practicable for other type of sensor and other similar systems. The system is simple and suitable for engineering use and has its practical value.
出处
《传感技术学报》
CAS
CSCD
北大核心
2007年第4期816-819,共4页
Chinese Journal of Sensors and Actuators
关键词
非线性特性
压力传感器
非线性校正
小波神经网络
遗传算法
nonlinear characteristic
pressure sensor
nonlinear emendation
wavelet neural network
genetic algorithm
