Sampling Method And Implementation Of Multichannel Ultrasonic Synthetic Signal Based On Compressed Sensing

Ultrasonic sensor array structure can effectively improve the inspection efficiency in the process of material defect detection,so it has a wide range of applications in the field of ultrasonic nondestructive testing.Array structure of ultrasonic sensor number,however,in the long time or long distance detection,if according to the conventional Nyquist sampling method for data collection,need to be greater than or equal to two times the effective bandwidth of the ultrasonic signal sampling frequency,the resulting huge amount of data for subsequent transmission,storage,and processing difficulties,seriously restrict the development of ultrasonic testing technology.Compressed Sensing is a sub-Nyquist sampling theory developed in recent years,which can effectively reduce the amount of data acquisition,but the current research on Compressed Sensing theory in the field of ultrasound is mainly aimed at the single channel ultrasonic signal,there is no research on the theory and method of multichannel synthetic signal Compressed Sensing.In view of this situation,this thesis firstly synthesizes the received multi-channel ultrasonic detection signal at the analog end to reduce the number of channels of the analog signal,and then the composite signal is sampled by compressed sensing,and the composite signal waveform is recovered from the compressed sensing sample data.,and separates each single channel parameter from it.Through the combination of the two channels,the purpose of reducing the amount of detection and collection data is achieved.In this thesis,according to the characteristics of pulse ultrasonic echo signal and the principle of ultrasonic detection,a multi-channel ultrasonic signal synthesis method based on the idea of frequency division multiplexing is proposed,and a signal synthesis circuit is designed and fabricated to realize the synthesis of the detection signal at the analog end.Combined with the principle of compressed sensing,the synthetic signal was sampled by compressed sensing,and the synthetic signal was reconstructed by orthogonal matching tracking algorithm.Finally,the single channel signal parameters were separated by digital filter method.The proposed method is verified by experiments with pipeline defects as the detection object.The results show that the proposed method can effectively reduce the amount of data,and can accurately calculate the amplitude and delay information contained in each single channel signal from the collected data.The main research contents and conclusions of this thesis are as follows:(1)The research status of multichannel ultrasonic signal synthesis methods,the research status of sampling theory and the research progress of compressed sensing theory are analyzed at home and abroad,and their characteristics are summarized.(2)The principle of pulsed ultrasonic detection is described,and the synthesis method of multi-channel ultrasonic signal based on the idea of frequency division multiplexing is proposed,and the mathematical model of multi-channel ultrasonic synthesis signal is established,and then the feasibility of the synthesis method is verified.The sparsity of dual-channel and three-channel ultrasonic synthetic signals is analyzed through simulation experiments.The results show that the synthetic signals meet the precondition of the application of compressed sensing theory.(3)According to the characteristics of the ultrasonic echo signal and the synthesis method,a limiter circuit,echo signal amplifier circuit,denoising circuit,post amplifier circuit,envelope extraction circuit,modulation circuit,bandpass filter circuit and inphase addition circuit are designed.The multi-channel ultrasonic signal receiving and synthesizing circuit is composed,and the circuit is tested and tested,and the circuit performance meets the requirements.(4)The basic principle of compressed sensing theory is analyzed,the compressed sensing method of multi-channel ultrasonic composite signal is proposed,and the compressed sensing and reconstruction performance of composite signal under different measurement matrices is studied.The simulation results show that the reconstruction precision of some orthogonal matrices is better than other measurement matrices.(5)A digital filter is proposed to separate the compressed sensing reconstructed synthetic signals.In the simulation test,the amplitude and time delay information of each single channel signal are accurately calculated by using the separated data,which verifies the feasibility of this method.(6)The proposed method was verified by setting up a test platform with pipeline defects as the detection object.The system performance was analyzed with the data collection volume,synthesized signal reconstruction accuracy,and the resolution accuracy of each single channel amplitude and delay parameters as indicators.The results show that compared to the conventional sampling method for the three-channel signals,the overall data volume is reduced by 87%,the relative reconstruction error of the ultrasonic synthesized signal is less than or equal to 7.13%,and the relative delay error of each single channel after separation is less than or equal to 1.71%,the relative amplitude error is less than or equal to 13.54%.