摘要
压力传感器的输出随温度呈非线性变化,同时含有较大的随机噪声。针对BP神经网络对压力传感器温度补偿建模时误差较大的问题,提出了基于灰色模型和BP神经网络的压力传感器温度补偿模型。首先,用灰色模型对数据进行预处理,以减小原始数据的噪声;然后,用降噪后的样本数据作为BP神经网络的输入进行训练。在相同的训练次数下训练误差可减小一个数量级。结果表明,采用该模型补偿后的压力传感器补偿精度明显优于BP网络模型。
The output of the pressure sensor exhibits a nonlinear change with varying temperature and contains significant stochastic error. In view that the BP（Back Propagation） network model makes it extremely difficult to realize high precision compensation, a hybrid model based on grey model theory and 13P neural network is put forward. First, it pre-processes the gyro output using the grey model to reduce the noise. Then it uses the processed sample data to train the BP network, so the training errors are reduced by one magnitude within the same training times. The result indicates that the compensation accuracy is improved greatly, which is obviously Superior to the BP network model.
出处
《半导体光电》
CAS
CSCD
北大核心
2013年第4期709-712,共4页
Semiconductor Optoelectronics
基金
国家自然科学基金项目(61204127)
黑龙江省教育厅2013科学技术研究项目(12531762)
黑龙江省普通高校新世纪优秀人才培养计划项目(1253-NECT025)
齐齐哈尔大学青年教师科研启动支持计划项目(2012k-M10)
关键词
灰色模型
BP神经网络
压力传感器
温度补偿
grey model
BP neural network
pressure sensor l temperature compensation
