Research On Speed Signal Detection And Frequency Algorithm Based On Doppler Effect

With the increasing maturity of LDV technology,it has been widely used in foreign countries,such as fluid velocity measurement,gas velocity in pipelines,blood flow velocity in blood vessels,military navigation,car navigation systems,and so on.In comparison,domestic development is relatively slow,mostly staying in the laboratory stage,few engineering products can be practical,and mainly focus on Doppler velocity measurement theory,optical path structure,signal detection and other aspects of research.One of the key issues in the design of LDV systems is the high-precision speed signal frequency solution.Based on the above analysis,the dissertation mainly studies the detection and frequency solving algorithms of solid motion laser Doppler velocimetry signals.By establishing an LDV speed measurement system platform,the speed simulation and experimental research are carried out.According to the Doppler effect principle of light,the advantages and disadvantages of three kinds of optical path models: optical heterodyne detection reference light model,single beam-double scattering model and dual beam-double scattering model are discussed.The mechanism of the variable frequency signal generated by the moving particle cutting stripe model is used to establish a mathematical model of the laser Doppler velocity measurement of the solid surface scattering motion particles.Based on this,a solution algorithm for laser Doppler signal frequency is proposed.The signal spectrum information is obtained by FFT spectrum analysis,and the analysis error caused by spectrum leakage,spectrum aliasing and fence effect generated by spectrum analysis is used to obtain the refined spectrum by Zoom-FFT spectrum refinement,and the error phenomenon is reasonably suppressed,and then the window function is adopted.The interpolation algorithm further performs error suppression and performs frequency adjustment.The three-spline fitting function is used to quickly fit the interpolation coefficient,improve the operation speed,obtain the optimal frequency spectrum information,and obtain the high-precision speed measurement signal.On the basis of the frequency solving algorithm,the simulation data and the experimental data are separately analyzed by MATLAB software.The frequency is solved in the low frequency range of 4MHz and the high frequency range of 4GHz during the simulation.In the LDV system,a DC motor is used to drive a metal disk with a diameter of 150 mm to simulate a moving body such as a steel strip of a rolling mill,and the rotational speed of the disk is measured by a photoelectric encoder,and the reference linear velocity is calculated by the rotational speed and the diameter of the disk.The operating system then performs frequency calculation on the experimental signal with a sampling rate of 10 MHz.The results show that the accuracy of simulation error and the accuracy of experimental error are better than 0.2%.However,by comparing the signal simulation curve with the experimental curve,the experimental curve fluctuates greatly,which indicates that the actual noise is greater than the signal simulation noise,and the noise in the measurement signal is large,which affects the fluctuation of the frequency of the experimental detection signal.Therefore,a series of interference noises interposed in the laser Doppler velocimetry signal are studied by a new signal denoising method.It is to use the characteristics of wavelet to select the method of zeroing the high frequency coefficient to denoise.The simulation and experimental results show that the high-frequency and low-frequency parts of the signal are ideally recovered,the signal-to-noise ratio is improved and a smooth signal is obtained.