Research On Ultrasonic Inner Inspection Technology Of Pipeline Corrosion

As one of the important methods for inspecting the pipeline corrosion, ultrasonic inner inspection can detect and position the wall defects on-line, and provide the basis for pipeline repair or abolition. Not only the malignant accidents, such as oil leak, pipe burst and so on, can be avoided, bankroll can be saved, but also the production efficiency can be improved. In the ultrasonic inspection, the mass echo data from the probes must be stored within instant time because of the higher pulse repetition frequency. Moreover, the useful signals are recognized difficultly duing to the operating mode complexity of the actual pipeline. Therefore, the echo signal processing is the key technology of ultrasonic inner inspection on pipeline corrosion. At present, the main problems include the real-time online storage of A-wave mass data has not been solved effectively, the accuracy of the B-scan conversion algorithm for the wall thickness is lower and the misjudgment rate and miss rate are higher, the post-processing work of C-scan image is inadequate, and the offline analysis software suitable for our national conditions is lacking.Aiming at the above difficulty, the engineering detection environment of pipeline and structure characteristics of the inner detector were simulated, and a multi-channel ultrasonic inner inspection experiment system for pipeline corrosion was built. The efficient compression and real-time storage of A-scan mass data from the test block and the standard pipeline, the accurate and automatic formation of B-scan thickness data and the defect identification of C-scan image data were studied deeply. The offline analysis software of the ultrasonic inner inspection on pipeline corrosion was developed. The specific research works are as follows:1. According to the technologic circumstance of the detected pipeline, the corresponding performance parameters of the ultrasonic probes were put forward, and on this basis the installation and operation mode of 24 probes was designed. The ultrasonic probes and the support frames of probe group as the main body of the ultrasonic inspection subsystem was built. The data acquisition, compression and storage subsystem was finished. With the computer subsystem, the walking and driving subsystem, and the power subsystem, a multi-channel ultrasonic inner inspection experiment system for pipeline corrosion was built.2. The forming principle of A-wave data was analyzed in detail and the compression algorithm of A-wave data based on the peak point extraction combined with the threshold segmentation was proposed. The primary data compression was based on the threshold segmentation algorithm. The secondary data compression was based on the peak point extraction algorithm. The validity of the compression algorithm was verified by the traditional data compression index and the thickness calculation. The results showed that the compression factor could achieve 200-300, the algorithm is simple and the computing speed is fast under the condition of ensuring the accuracy of thickness measured. This algorithm could be realized only using a piece of FPGA on each card, which solved effectively the online real-time compression and storage problem of mass data from the pipeline inspection.3. The Fast Fourier Transform (FFT) had advantages of simple principle and fast operation used commonly in the wall thickness conversion algorithm, but the accuracy of conversion is not high and it is easy to misjudge. In order to solve this problem, the 11/2-dimensional spectrum estimation of higher-order statistics was adopted to process the ultrasonic echo signals. According to the periodic characteristics of the spectrum figure, the secondary 11/2-dimensional spectrum estimation algorithm was adopted. On the basis of the wall thickness frequency ranges, the improved algorithm based on the secondary 11/2-dimensional spectrum estimation was put forward. Studies showed that the algorithm highlighted the transit time value of the wall thickness inspection and improved effectively the automatic extraction accuracy of the wall thickness feature. This algorithm, the relative error of which was less than 3%, was better enormously than the FFT algorithm.4. C-scan image was reprocessed preliminarily adopting the median filtering method, the impulse noise is removed effectively that made the image more smooth, clear and intuitive.5. According to the needs of the offline analysis and the visual display of the pipeline detection data, the offline analysis software for the ultrasonic inner inspection system of pipeline corrosion was developed. This software could display clearly the A-scan waveform and spectrum of any detected point within 24 channels, B-scan waveform of any channel, and C-scan images of the whole detected area. Through the selection of the channel number and the detected point in the main interface, the users may obtain the corrosion status of the pipeline simply and quickly.Through a lot of theoretical analysis and experimental research, the key technical problems, such as efficient compression and real-time storage of mass data from pipeline ultrasonic inner inspection, signal reduction and processing, and defect recognition were solved effectively. It is great significant to extend the service life and guarantee the safe operation of pipeline.