The Hardware Implementation Of Ultrasonic Digital Signal Processing Algorithm And The Development Of New Digitized Ultrasonic Flaw Detectors

This report is the total summary of the research work during the period in which I have been engaged in the postdoc R&D working station, Guangdong goworld Co.,Ltd. The main achievements have been made are as follow as: The software based ultrasonic signal processing algorithm which is conventionally implemented using software programming is transferred,to integrated hardware implementation based on the programmable application specific integrated circuits(ASIC). A novel layout of hardware and software design for the digitized ultrasonic flaw detector is proposed , which is based on the PC and FPGAs embedded modules. Also a kind of digitized ultrasonic flaw detectors which is used specifically by railway locomotive flaw detection are developed. A noise reduction method to the ultrasonic signal is proposed using the discrete wavelet transform. The multiresolution analysis for reducing the noise in ultrasonic signal is performed and its orthogonal filter is designed.In order to improve the performance of the digitized ultrasonic flaw detector with heightening the frequencies of ultrasonic wave transmitting and displaying, some digital signal processing algorithms and logical circuits in traditional ultrasonic flaw detectors are substituted with an application specific IC which is designed on FPGAs technology. Firstly, the principle of digitized ultrasonic flaw detector is discussed, The method of ultrasonic signal sampling and data acquisition and compression and gates for alarming are presented. Then a simple algorithm of wave displaying data arrays forming which is applicable to hardware implementation is proposed according to the working character of LCD driver. The digital logic circuits performing the functions of digital demodulation, sampling control, data compression, gates setting and two kinds of wave displaying data arrays are designed according to the characters of FPGAs. Lastly, they are transferred to netlist file which will be downloaded to the Xilinx Spartan XCS320XL after simulation, analysis, synthesis through FPGAs development tools Foundation Express 2.1. The hardware implementation of ultrasonic signal processing is applied to the ultrasonic flaw detector with which the performances are greatly improved.According to the requirements of railway locomotives flaw detection, a new design scheme for the ultrasonic flaw detector is proposed based on the PC104 and FPGA embedded modules structures which can provide a more flexible, strong adaptable, extendable hard ware platform with a fast signal processing ability. Theoverall hardware scheme and the choice of modules are discussed in detail. A programmable gain control circuits with a wide range of gain variation is constructed using the ultra low noise devices. The programmable amplifier AD604, the analog to digital data converter AD9012, the digital to analog converter AD7840 are discussed insightfully in their principle and characters. The overall object oriented and modulated software programming method is used for the specific flaw detection. The detail flow charts to all modules functions programming are described in the report. The developed sample instrument is debugged and has been verified to meet the requirements of locomotive flaw detection.The discrete wavelet transform based multiresolution analysis to ultrasonic signal is discussed for the purpose of noise reduction. From the engineering analysis and application point of view the principles of continuing wavelet transform and the wavelet transform on discrete grids and the design methodology of the orthogonal FIR filters, which perform multiresolution analysis and synthesis to the discrete ultrasonic signals, are analyzed in detail. The daubechies wavelet FIR filters is induced to ultrasonic signal processing system through analyzing the ultrasonic noise signal models. All the simulation results verified the good performance of the discrete wavelet transform application to the ultrasonic signal processing systems. Its implementation is also applicable to practical ultrasonic detection instruments.