摘要
提出基于BP神经网络的浓度传感器非线性误差校正方法。文中详细给出了BP神经网络算法原理及训练方案。当替换传感器或环境条件发生变化时,只要获取一组输入输出样本对,便可重新训练网络,获得新的输入输出样本关系,从而实现传感器非线性校正和动态标定,提高传感器的互换性,有实际应用价值。
The method for correcting the sensors' nonlinear error is presented based on the BP neural network.It is detailed to introduce the algorithm principle and trains project of the BP neural network.When the replacement of transducer or environment occurrences variety,as long as obtaining a group of input and output samples,the network can be retrained,acquire the new input and output samples relations to.Thus,the correction of nonlinear error is realized,and it can increase transducer compatibility and has practical applied value.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第3期226-228,共3页
Computer Engineering and Applications
基金
辽宁省教育厅科学研究计划(编号:202023083)资助
关键词
非线性校正
BP网络
浓度传感器
nonlinear correction,BP neural network,concentration transducer
