【作者】 刘新宇；

【导师】 米建伟；

【作者基本信息】 西安电子科技大学 ， 机械电子工程， 2023， 硕士

【摘要】 飞机电缆是飞机动力系统和控制系统的重要装置,分布在飞机中各个位置,具有数量多、结构复杂等特点,对飞机安全运行起到了关键作用。一旦电缆出现故障没有及时发现并解决,可能会造成灾难性的后果。因此,针对飞机电缆故障进行在线检测研究对保障飞机安全有着重要意义。本文研究基于行波反射原理的电缆故障检测方法,采用单端检测的方式对电缆故障进行定位,研究的主要内容如下:首先,介绍了电缆故障检测技术的国内外研究现状和现有的故障检测方法,建立了电缆的分布式参数模型,研究了行波在传输线中的波过程,对行波的反射与透射现象进行了分析,在此基础上将调制解调机制与扩展频谱时域反射法(SSTDR)相结合,进一步研究可用于在线检测的检测方法,对测距指标、测试信号的选择等做了详细分析,为后续研究打下基础。接着,针对在低信噪比下故障检测精度不足的问题,对基于SSTDR方法的时延估计算法进行改进,在二次互相关算法的基础上进行加权,并引入最小二乘拟合方法对相关函数波形的峰值点进行锐化处理,结果显示在低信噪比较低时具有良好的时延估计效果。同时,为了保证系统检测的实时性,提出使用稀疏快速傅里叶变换(SFFT)对时延估计算法进行优化,对故障信号进行重排、平滑滤波以及降采样处理,降低算法的复杂度。通过对时延估计标准差的分析,实现了较为精准的时延估计。最后,利用Simulink仿真平台搭建检测系统模型,并利用MATLAB完成核心算法。仿真结果表明,本文研究的时延估计算法能够在低信噪比的条件下准确地对电缆故障进行诊断定位。设计电缆故障在线检测系统,系统采用ARM和FPGA双处理器架构,设计了测试信号发射电路、信号采集电路、信号调理电路等关键电路,完成ADC数据采集、测试信号生成等FPGA自定义逻辑开发,以及Nios II的主函数和UART串口通信函数的编写。实现了测试信号输出、信号的采集调理、波形数据显示等功能。并利用Lab VIEW软件设计电缆故障检测系统的上位机,搭建电缆故障检测的实验平台,对检测系统进行实验测试,验证了该系统能够实现故障定位的功能。更多还原

【Abstract】 Aircraft cable is an important device of aircraft power system and control system,distributed in various positions in the aircraft,with a large number and complex structure,etc.,which plays a key role in the safe operation of the aircraft.Since the existing offline detection methods cannot detect aircraft cables in real time,it is of great significance to conduct online detection research on aircraft cable faults to ensure aircraft safety.In this thesis,the cable fault detection method based on traveling wave reflection is studied,and the single-ended detection method is used to locate the cable fault,and the main contents of the research are as follows:Firstly,the research status of cable fault detection technology at home and abroad and the existing fault detection methods are introduced.The distributed parameter model of the cable is established,the wave process of the traveling wave in the transmission line is studied,and the reflection and transmission phenomena of the traveling wave are analyzed.On this basis,drawing on the modulation and demodulation mechanism in the field of communication,the principle of spread spectrum time domain reflectometry(SSTDR)that can be used for online detection is further studied,and the ranging index and test signal selection are analyzed in detail,which lays a foundation for subsequent research.Then,aiming at the problem of insufficient fault detection accuracy under low signal-tonoise ratio,the delay estimation algorithm based on SSTDR method is improved,weighted on the basis of quadratic cross-correlation algorithm,and the principle of least squares fitting is introduced to sharpen the peak point of the relevant waveform,and the results show that it has a good delay estimation effect when the low signal-to-noise ratio is low.At the same time,in order to ensure the real-time detection of the system,it is proposed to use sparse fast Fourier transform(SFFT)to optimize the delay estimation algorithm,and perform frequency basketing,downsampling and reconstruction processing of fault signals to reduce the complexity of the algorithm.Through the analysis of the standard deviation of delay estimation,a more accurate delay estimation is realized.Finally,a detection system model was built using the Simulink simulation platform,and the core algorithm was completed using MATLAB.The simulation results show that the delay estimation algorithm designed in this thesis can accurately diagnose and locate cable faults under low signal-to-noise ratio conditions.Design and implement an online detection system for cable faults.The system adopts a dual processor architecture of ARM and FPGA.Key circuits such as test signal transmission circuit,signal acquisition circuit,and signal conditioning circuit are designed.FPGA custom logic designs such as ADC data acquisition and test signal generation are completed,as well as the main function and UART serial communication function of Nios II are written.Implemented functions such as testing signal output,signal acquisition and conditioning,and waveform data display.And the upper computer of the cable fault detection system was designed using Lab VIEW software,and an experimental platform for cable fault detection was built.The experimental testing of the detection system verified that the system can achieve fault location function.更多还原