机械振动的超声波检测技术研究

发布时间：2018-03-12 18:53

本文选题：声场　切入点：振动检测　出处：《华东交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：非接触式振动检测方法是一类重要的机械振动检测方法,其在振动检测领域中起到无可替代的重要的作用。在非接触式振动检测方法中,传感器与被测物体相互隔离,能够在不干扰被测对象的情况下实现振动检测,尤其可用于对表面粗糙、细微、质轻及柔性结构的振动检测。在常见的非接触式振动检测方法中,基于超声波的机械振动检测方法即超声波测振法有其自身的独特优势,具有测量精度高、测量电路简单、成本低、通用性强等特点,能够运用于烟雾、不透明气体、强腐蚀、强光等环境中进行振动检测,对超声波反射性能好、表面粗糙度小于超声波波长的表面都能对其进行振动检测。因此,有必要对超声波测振技术进行深入研究。本文首先对超声波振动检测技术及超声波测振信号解调算法进行了综述;基于超声波的多普勒效应及声参量效应详细介绍了超声波测振原理,并建立了受振动调制的超声波反射回波数学模型;根据超声波反射回波数学模型以及超声波振动检测原理得出了影响超声振动检测精度的三个主要因素—超声波声速、入射超声波频率、传感器探头安装位置。对超声波声速的影响因素以及传感器的安装位置进行了深入研究,提出了修正超声声速的方法,确定了传感器的合理安装位置。参考前人的研究成果,综合考虑不同频率超声波在空气中的传播特性,选择采用40KHz的超声波信号进行振动检测研究;以中心频率为40 KHz超声波传感器直径大小为研究对象,对不同尺寸的传感器进行声场及指向性研究,基于该研究结果,结合成本因素及传感器获取的难易程度,对超声波传感器进行选型。采用Hilbert变换解调算法及能量算子解调算法分别对超声波测振仿真信号进行解调,提取出被测物体的振动速度信号。针对选用Hilbert变换解调算法获取的振动速度曲线有端点效应,能量算子解调算法因误差项的存在限制了所能测量的振动频率范围从而影响解调精度问题,提出了一种改进的归一化复域能量算子解调算法并用该算法对超声波测振信号进行处理,将处理结果与Hilbert变换解调算法及能量算子解调算法的解调结果进行对比分析,结果表明该方法能获得更好的解调效果。最后,基于上述研究进行超声波振动检测系统设计,搭建超声振动检测实验台,开展超声波振动检测实验,将超声波测振信号解调得到的振动速度与加速度传感器输出信号积分求得的振动速度信号进行对比分析,实验结果证明了上述理论研究的正确性和可行性。
[Abstract]:Non-contact vibration detection method is an important kind of mechanical vibration detection method, which plays an irreplaceable role in the field of vibration detection. It can be used to detect vibration without disturbing the object, especially for rough, fine, light and flexible structure. In the common non-contact vibration detection method, Ultrasonic vibration detection method based on ultrasonic has its own unique advantages, such as high measuring precision, simple measuring circuit, low cost and strong generality. It can be used in smoke, opaque gas, strong corrosion, etc. Vibration detection in strong light and other environments can detect the vibration of ultrasonic wave when the surface roughness is less than ultrasonic wave length. It is necessary to study the ultrasonic vibration measurement technology deeply. Firstly, the ultrasonic vibration detection technology and the demodulation algorithm of ultrasonic vibration signal are summarized in this paper. Based on the Doppler effect and parametric effect of ultrasonic wave, the principle of ultrasonic vibration measurement is introduced in detail, and the mathematical model of ultrasonic reflection echo modulated by vibration is established. According to the mathematical model of ultrasonic reflection echo and the principle of ultrasonic vibration detection, three main factors affecting the accuracy of ultrasonic vibration detection are obtained: ultrasonic velocity, ultrasonic frequency, and ultrasonic frequency. The influence factors of ultrasonic sound velocity and the installation position of sensor are studied in depth, and the method of correcting ultrasonic sound velocity is put forward, and the reasonable installation position of sensor is determined. Considering the propagation characteristics of ultrasonic wave with different frequencies in the air, the ultrasonic signal of 40kHz is selected for vibration detection, and the diameter of the ultrasonic sensor with central frequency of 40kHz is taken as the research object. The acoustic field and directivity of sensors with different sizes are studied. Based on the results of the research, combined with the cost factors and the degree of difficulty in obtaining the sensors, Hilbert transform demodulation algorithm and energy operator demodulation algorithm are used to demodulate the simulation signal of ultrasonic vibration measurement. The vibration velocity signal of the measured object is extracted. The vibration velocity curve obtained by using the Hilbert transform demodulation algorithm has an endpoint effect. The energy operator demodulation algorithm limits the range of vibration frequency which can be measured because of the existence of error term, which affects the demodulation accuracy. An improved normalized complex energy operator demodulation algorithm is proposed to process ultrasonic vibration signal. The results are compared with those of Hilbert transform demodulation algorithm and energy operator demodulation algorithm. The results show that this method can obtain better demodulation effect. Finally, the ultrasonic vibration detection system is designed based on the above research, and the ultrasonic vibration detection experiment bench is built, and the ultrasonic vibration detection experiment is carried out. The vibration velocity obtained by demodulating the ultrasonic vibration signal is compared with the vibration velocity signal obtained by the integral of the output signal of the acceleration sensor. The experimental results prove the correctness and feasibility of the above theoretical research.
【学位授予单位】：华东交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH113.1;TB559

【参考文献】