摘要
为提高智能传感器的测量准确度,利用神经网络良好的非线性映射能力,对传感器的标定数据进行输入-输出特性的反非线性逼近,同时,利用传感器实验数据进行神经网络的训练。结果表明:与传统的数据处理方法相比较,利用神经网络进行的传感器数据处理,能使传感器的准确度由±6.67提高到到±0.98。
The nonlinear reflecting capability of neural network is applied to approach the I/O characteristic curve to improve the accuracy of measurement of the sensor,the sensor can get a I/O calibration with higher accuracy,and at the same time it processes intelligent by extracting information from raw data through the neural network.The result indicates that the accuracy of measurement can be increased from ±6.67 % to ±0.98 % compared with traditional data processing.
出处
《传感器技术》
CSCD
北大核心
2004年第8期52-54,共3页
Journal of Transducer Technology
关键词
神经网络
智能传感器
数据处理
neural network
intelligent sensor
data processing
