Research On Low Frequency Compensation And Self-noise Evaluation Of Magnetoelectric Vibration Sensor

Vibration is very common in life,such as the vibration of large rotating machinery,the vibration caused by vehicle driving,the natural vibration of railways,bridges and buildings,etc.The frequency of this type of vibration is low,generally below 10 Hz.Magnetoelectric vibration sensor is a kind of vibration sensor with low frequency band.However,the existing magnetoelectric vibration sensor has a relatively high natural frequency,and its low-frequency measurement lower limit is difficult to meet the requirements of low-frequency measurement.Therefore,its low-frequency characteristics need to be compensated.At the same time,the evaluation of the performance of the magnetoelectric vibration sensor needs to analyze its own noise level.The self-noise level is also a prerequisite for real-time determination of whether the vibration sensor is degraded and reasonable use.However,the actual measurement environment contains a lot of external noise interference.It is very necessary to study how to evaluate the self-noise level of the vibration sensor in a noisy environment.In this thesis,a low-frequency compensation method of magnetoelectric vibration sensor is studied,and the method of evaluating the self-noise level of magnetoelectric vibration sensor in a noisy environment is studied.The main research contents are as follows:1.The frequency response characteristics of the magnetoelectric vibration sensor are studied.In this thesis,the magnetoelectric vibration sensor is firstly modeled,and the transfer function of the system is comprehensively derived;secondly,the correctness of the model is verified through experiments;finally,the frequency response characteristics of the magnetoelectric vibration sensor under different sensitive parameters(velocity and acceleration)and different parameters(natural frequency,sensitivity and damping ratio)are analyzed in depth.The analysis shows that the response characteristics of the magnetoelectric vibration sensor become worse with the decrease of frequency below the natural frequency.The response characteristic of the electromagnetic induction system is the reason for the insufficient low-frequency response characteristic of the magnetoelectric vibration sensor.2.A low-frequency compensation method of force balance negative feedback is studied.Based on the comparative analysis of the respective characteristics of passive correction method,series correction method,feedback correction method and digital filter correction method,the feedback correction method is selected for low-frequency compensation of magnetoelectric vibration sensor.First,the low-frequency compensation network is designed according to the principle of force balance negative feedback;secondly,the frequency response characteristics and stability of the compensation network are simulated and analyzed;finally,the low-frequency compensation performance of the compensation network is verified through experiments.The results show that the low-frequency compensation method of force balance negative feedback can reduce the natural frequency of the magnetoelectric vibration sensor from 4.5 Hz to 0.6 Hz.3.The method of evaluating the self-noise of magnetoelectric vibration sensor under noisy environment is studied.This thesis firstly uses single-channel self-noise measurement,two-channel coherence analysis,and three-channel coherence analysis to measure,evaluate and compare the self-noise of magnetoelectric vibration sensors in low-noise and high-noise environments.Secondly,this paper introduces the multi-channel coherent subtraction method to the self-noise evaluation of the magnetoelectric vibration sensor,and verifies the effectiveness of the method to remove the environmental interference noise and obtain the self-noise through the simulation experiment.Finally,experiments verify the effectiveness of multi-channel coherent subtraction in evaluating self-noise The results show that the multi-channel coherent subtraction method can effectively remove environmental interference and noise,and the more reference channels used,the smaller the residual output.It is a very effective method for real-time monitoring of the self-noise level of the magnetoelectric vibration sensor in a noisy environment.