| Because of their own limitations of the conventional thickness measurement methods, it is difficult to measure the thickness of the ultra-thin metal material that less than 0.1mm accurately. So these methods can’t be able to meet the requirements of modern industry for nondestructive testing, like high reliability, high precision and high efficiency and so on. With the development of the embedded systems and computer technology, we can implement a digital ultrasonic thickness measurement system use of the chips like FPGA, DSP and ARM. Through the system, we can transfer the experimental data to computer quickly and accurately. This data is a non-stationary signal, so we can’t accurately describe its time-frequency characteristics simply by Fourier transform. We also need the help of time- frequency analysis technology for further processing.In this thesis mainly includes three aspects: half-wave resonance method, hardware circuit design and time-frequency analysis technology.First,the main study of this thesis is the ultra-thin metal material within 0.02 mm ~ 1mm. Conventional ultrasonic thickness measurement methods have blind, unable to complete the survey work.Therefore, on the basis of conventional ultrasonic thickness measurement methods and experimental analysis and verification, measure the thickness of the ultra-thin metal by the half-wave resonance method. Half-wave resonance method is an extension application of the resonance thickness measure theory. We can achieve the ultra-thin material thickness measurement experiments at a lower ultrasonic excitation frequency. Moreover, the higher of the frequency, the more accurate of the measurement results we can get.In this design, the range of the sweep pulse frequency is 5MHz ~ 15 MHz. Choice apposite ultrasonic transducer that made by new composite material, its frequency can meet the needs of the experiment.Second, design the hardware circuit to collect the experimental data. For ultrasonic signal 5MHz ~ 15 MHz, the sampling rate of the A/D chip is 80 MSPS, ensuring the completeness of the signal acquisition.We make an integrated design with FPGA and ARM, to achieve the fast and efficient signal acquisition system. They both has their own advantages and cooperate with each other, make the FPGA as the core processor, ARM auxiliary control A/D acquisition and Ethernet communications. Through RJ45 interface and network cable to achieve the data transfer with the computer. Further, design the signal receiving circuit to amplification the weak ultrasonic signal, and weaken the noise. There are protection circuit and bias circuit in the signal receiving circuit, to ensure the safety and match with subsequent circuits.The interaction between each stage circuit is also noted, it contains both analog and digital circuit, and work in a high frequency environment. In the circuit design and PCB drawing process, we take appropriate measures to minimize the interference caused by components.Third,it is hard to accurately determine the position of resonance frequency by observe the time domain waveform and its Fourier transform of the experimental thickness measurement data in MATLAB. Therefore, we compare the commonly used frequency analysis methods, like Short-time Fourier transform, wavelet analysis and Wigner-Ville distribution and use Wigner-Ville distribution to deal with the data because of its super Time-frequency aggregation.We can see the scan frequency have a linear trend over time from the time domain clearly. While according to the energy values, draw three-dimensional map of WVD. Accurate positioning of the maximum energy value in it, the center frequency can be accurately determined to obtain more accurate results.Finally, we can obtain comparatively accurate measurement of the thickness by analyze the experimental data for the different thickness of the metal sheet. At the same time, verify the feasibility of this hardware circuitry and the half-wave resonance method. And for the thinner material, the absolute error is smaller. |
PDF Full Text Request