摘要
利用误差反向传播算法(BP)和径向基函数(RBF)神经网络法,分别对医用体内光纤温度传感器探头进行设计,训练神经网络的数据通过实验和插值得到。该方法具有准确、可靠和知识辅助设计的特点。结果证明,此方法可以节省设计成本,缩短设计周期,有很高的实用价值。
BP and RBF neural network method has been used to design optical fiber temperature sensor probe using in medical treatment. Datum trained for neural network is gained from experiment and interpolation. This method has the characteristic of accuracy,credibility and knowledge-aid design. It can save the cost of device design and decrease period of design. This method has a good prospect of application.
出处
《压电与声光》
CSCD
北大核心
2004年第2期92-94,共3页
Piezoelectrics & Acoustooptics
关键词
光纤
温度传感器
建模
神经网络
optical fiber
temperature sensor
modeling
neural network
