论文部分内容阅读
分振幅光偏振仪是使用范围十分广泛的测量传感器,为了解决分振幅光偏振仪数据过程中存在的精度低难题,提出了基于神经网络的分振幅光偏振仪数据处理方法。首先收集分振幅光偏振仪的输入信号和输出信号,将它们组合在一起作为神经网络的学习样本,然后采用神经网络分振幅光偏振仪的学习样本进行拟合,进行分振幅光偏振仪数据处理,最后采用应用实例对该方法的有效性进行分析,结果表明,该方法提高了分振幅光偏振仪数据处理精度,而且测量效果要优于其它方法。
The polarization amplitude polarimeter is a measuring sensor with a very wide range of applications. In order to solve the problem of low precision in the process of polarization amplitude polarimeter, a data processing method of polarization amplitude polarizer based on neural network is proposed. Firstly, the input signal and the output signal of polarization polarizer are collected and combined together as a learning sample of the neural network. Then, the learning samples of the neural network polarization polarimeter are used for fitting, and the polarization amplitude polarizer data processing Finally, the effectiveness of this method is analyzed by application examples. The results show that this method can improve the data processing precision of polarization polarimeter and the measurement result is better than other methods.