基于BP网络的压力传感器信息融合

Information Fusion of Pressure Sensor Based on BP Network

压力传感器输出特性容易受环境温度、电压扰动等各种非目标参量的影响 ,从而大大降低了其性能。BP算法是一种最速下降的静态寻优算法 ,而对其改进的算法LMBP算法克服了标准BP算法的固有缺点 ,不但学习速度快 ,而且精度高。利用LMBP算法对压力传感器的输出进行融合 ,有效地消除了非目标参量特别是温度对压力传感器输出的影响 ,最后利用MATLAB软件对样本数据进行训练和仿真 ,通过对融合结果分析可知 :BP网络的LMBP算法不仅提高了压力传感器的精度 ,而且提高了压力传感器的稳定性和可靠性。

The output of pressure sensor is easily affected by non-objection parameters, such as environment temperatures, voltage fluctuation and so on, in the applications. BP algorithm is the steepest descent algorithm and an algorithm to improve its performance, LMBP algorithm overcomes the disadvantage of the standard BP algorithm and it has not only fast learning velocity, but also high accuracy. Using LMBP algorithm to fuse the output data of pressure sensor, it eliminates the affection of the non-objection parameters, especially, such as temperatures, to the pressure sensor. Then the MATLAB software is used to train and simulate the example data. Finally, the analyzed fusion results show that not only the accuracy of the pressure-sensor is improved, but also the stability and liability of the pressure sensors are improved.