GA-BP神经网络在气压式相对高度计中的应用研究

The Study on the Application of GA-BP Neural Network in the Pneumatic Pressure-Type Relative Altimeter

为了提高相对高度测量的精确性,研究并实现了一种基于气压传感器阵列式测量和遗传算法(GA)优化反向传播(BP)神经网络数据融合处理的高精度气压式相对高度计,给出了相应的硬件结构和软件设计。结合实验测量的数据和相关文献的数据,从准确性、稳定性和通用性的角度对GA-BP神经网络、传统BP神经网络以及标准计算公式在气压式相对高度计中应用的性能进行了对比分析。研究结果表明,本文提出的基于GA-BP神经网络的相对高度计具有更高的测量精度、更高的稳定性和更好的推广能力,能够满足日常相对高度的实时测量需求。

In order to improve the accuracy of the relative height measurement,a high-precision pneumatic relative altimeter has been studied and implemented through pressure sensor-array measurements and data fusion with an optimizing back propagation （ BP ） neural network algorithm based on genetic algorithm （ GA ） . The corresponding hardware and software designs have been provided as well. Combined with experimental measured data and relevant literature data,the application performances of the GA-BP neural network, the traditional BP neural network and standard formulas in the pneumatic relative altimeter were compared and analyzed in term of accuracy,stability and versatility. The results show that the proposed pneumatic relative altimeter based on GA-BP neural network has higher accuracy, higher stability and better ability to promote, and it can meet the daily needs of real-time measurement for the relative height.