BP神经网络辅助的氢气泄漏检测方法研究  

Hydrogen leakage detection method by using BP neural network

作　　者：姚璐峤 张小军[1] 王凯[1] 张蒙 张筱璐 李跃娟 苗扬[1,2] YAO Luqiao;ZHANG Xiaojun;WANG Kai;ZHANG Meng;ZHANG Xiaolu;LI Yuejuan;MIAO Yang(Faculty of Materials and Manufacturing,Beijing University of Technology,Beijing 100124,China;Beijing Key Laboratory of Advanced Manufacturing Technology,Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京工业大学材料与制造学部,北京100124 [2]北京工业大学先进制造技术北京市重点实验室,北京100124

出　　处：《重庆理工大学学报（自然科学）》2022年第3期289-294,共6页Journal of Chongqing University of Technology：Natural Science

基　　金：国家重点研发项目(2021YFB4001001);国家自然科学基金资助项目(51975011);北京工业大学国际科研合作种子基金项目(2021B24)。

摘　　要：氢能作为一种清洁高效的可再生能源,具有能量密度高、来源广泛、零污染等优点,被广泛认为是本世纪最具应用前景的能源载体之一。高压气态储氢是目前我国使用最为广泛的一种氢气储存方式,而高压氢气泄漏是高压储氢中的重大安全隐患。结合BP神经网络设计了一种检测高压氢气泄漏的方法。将激光束穿过氢气射流产生的光斑图像输入神经网络,从而反推出氢气泄漏口的直径和出口压力大小。结果表明:预测值与实际值接近,并且具有很高的稳定性。这项技术可以应用于检测远距离放置的储氢瓶阀门的泄漏、低压电解槽的泄漏、氢储罐管件及密封环的泄漏,以及储氢设备的通风口处的泄漏等。Replacing fossil energy with new energy is an important way to deal with carbon neutralization.As a clean and efficient renewable energy, hydrogen energy has the advantages of high energy density, wide sources and zero pollution.It is widely considered to be one of the most promising energy carriers in this century.High pressure gaseous hydrogen storage is the most widely used hydrogen storage method in China, and high-pressure hydrogen leakage is a major potential safety hazard in high-pressure hydrogen storage.Combined with BP neural network, a method for detecting high-pressure hydrogen leakage is designed in this study.The spot image generated by the laser beam passing through the hydrogen jet is input into the neural network, so as to deduce the diameter of the hydrogen leakage port and the outlet pressure.The results show that the predicted value is close to the actual value and has high stability.This technology can be applied to detect the leakage of the valve of the hydrogen storage bottle placed remotely, the leakage of the low-pressure electrolytic cell, the leakage of the pipe fittings and sealing rings of the hydrogen storage tank, as well as the leakage at the vent of the hydrogen storage equipment.

关 键 词：氢射流 泄漏检测 激光 神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]