GA-BP神经网络对SAW压力传感器测量数据的拟合  

Fitting analysis of SAW micro pressure sensor measurement data by ga optimized BP neural network

作　　者：许浩源 李媛媛[1] Xu Haoyuan;Li Yuanyuan(School of Electrical and Electronic Engineering,Shanghai University of Engineering Scienc,Shanghai 201620,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620

出　　处：《电子测量与仪器学报》2021年第4期7-14,共8页Journal of Electronic Measurement and Instrumentation

基　　金：国家自然科学基金(61803254)项目资助。

摘　　要：随着传感技术的不断发展,出现了越来越多以传感器为基础的无线传感检测系统。这些传感系统需要对采集到的数据进行数据分析。因此,传感器的数据分析对于无线传感系统的精确检测起到至关重要的作用。首先对设计的声表面波(surface acoustic wave,SAW)微压力传感器进行实际测量,利用最小二乘法建立数学模型,对测得的频率变化与对应的载荷力之间的关系进行数据拟合分析。随后,构建BP(back propagation)神经网络模型,通过该方法对已采集的数据进行样本训练并对SAW微压力传感器的输入与输出之间的关系进行预测。最后,使用遗传算法(genetic algorithm,GA)对BP神经网络进行优化。结果表明,经过优化后比优化前的BP神经网络误差减少近45%,从而验证了遗传算法能够优化BP神经网络的可行性。该分析方法对声表面波微压力传感器无线传感检测系统的发展提供了重要的研究依据。With the continuous development of sensing technology,more and more sensor-based wireless sensor detection systems have emerged.These sensor systems require data analysis of the collected data.Therefore,the data analysis of sensors plays a vital role in the accurate detection of wireless sensing systems.Firstly,the design SAW micro-pressure sensor is measured,and the mathematical model is established by the least square method,and the relationship between the measured frequency change and the corresponding load force is analyzed by data fitting.Subsequently,a BP neural network model was constructed to train the collected data and predict the relationship between the input and output of the SAW pressure sensor.Finally,genetic algorithm is used to optimize the BP neural network.The results show that after optimization,the error of BP neural network is reduced by nearly 45%compared with that before optimization,thus verifying the feasibility of genetic algorithm to optimize BP neural network.This method provides an important research basis for the development of wireless sensor detection system for acoustic surface wave micro-pressure sensor.

关 键 词：声表面波 传感器 最小二乘法 遗传算法 BP神经网络 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]