Study On High Capacity And Fidelity Ultrasonic Data Processing For Offshore Pipeline

Currently, offshore pipeline inspection technique is a research hotspot in the international nondestructive evaluation (NDE) field, but it is late for domestic research. The research on technologies, methods, and approaches for pipeline detection makes great theoretical and realistic sense. Experts and scholars attach importance to ultrasonic detection because of its merits.Based on the high technology research and development program"Offshore pipeline detection device and inspection technology", ultrasonic signal denoising, ultrasonic data compression, thickness of pipeline detection and pipeline defect recognition are studied deeply according to detection requirement of offshore pipeline. All these provide key technologies for the offline data analysis system of the program.The theory of ultrasonic thickness measurement and offshore pipeline detection device were introduced firstly. According to performance guideline of offshore pipeline detection device, the pipeline ultrasonic inspection device was designed, the parameters of the center frequency, the wafer diameter, SO were determined by analyzing and caculating. Ultrasonic sensors were selected and inspection transducer device was designed. The ultrasonic data storage system was developed.The one of key research on ultrasonic non-destructive detection is denoising ultrasonic echo signals. Ultrasonic RF signals are decomposed by empirical mode decomposition and IMFs of ultrasonic signal can be obtained. According to the performance of signal decomposition, the method of component selection based on energy is composed to denoise the ultrasonic signal. The experimental results and discussion demonstrated the usefulness and effectiveness of empirical mode decomposition as a signal processing technique for the analysis of ultrasonic waveforms. This method is adaptive. It is easy to be realized in computer. The multi-sensor data fusion algorithm has been applied to sensor array signal processing. A fusion method based on EMD and noise variance is proposed. Experiments in lab have testified the efficiency of this method. In addition, the comparison in fusion time and fusion results with existing fusion method based on wavelet and average technique shows the advantage of this method greatly. Collected and stored data are demodulation signal when thickness of offshore pipeline is detected by ultrasonic sensors array, signal noises represent as pulse noise. By analyzing the performance of pipeline detection ultrasonic signals, the ultrasonic signal is filtered by the median filter, a typical nonlinear filter. Experiments in lab have testified the efficiency of this method.The theory and method were introduced in this dissertation. Lifting wavelet transform has been used to ultrasonic signal compression. In order to obtain best compression ratio, some factors, such as selection a proper wavelet base, setting the decomposition level number, defining the threshold by wavelet coefficients and so on, must be considered. By analyzing the performance of pipeline detection ultrasonic RF signals, the best parameters were confirmed to get the good compression result. The parameter estimation methods of the continuous wavelet transform and maximum likelihood estimation were introduced in this dissertation. MP based GA is applied to estimate the parameters of atom functions that best represent the detected ultrasonic signal in this dissertation. GA is a numerical optimization method based on the concepts of genetics and natural selection. It is an efficient technique to optimize difficult functions in large search spaces. The simulation signals and experimental results have testified the efficiency of this method. Using LS, it also need spend considerable searching time and find the time-frequency atoms best matched original signal. Genetic algorithm (GA) can estimate the values of the parameter vector, in order to reduce the calculation amount. The collected data must be compressed and saved timely during the inspection. Ultrasonic detecting device requires lossless and real-time (high speed) data compression for the inspection of long time. The method of Huffman coding is selected by comparing some sorts of lossless compression methods.The algorithms of pipeline thickness detection based on FFT and correlation were expatiated, in order to enhance measurement precision, second FFT method was proposed. The Hilbert-Huang transform was used to extract the signal features of time domain, frequency domain and time frequency domain. The expert system was used to process the ultrasonic data of pileline, defect database was built and the classidication and grade of pipeline defects was deduced.Because of the offshore pipeline detection entironment variety, the flaws of the pipeline are multiplex. Some standard defect samples were made by experiment machining.Experiments showed that the accuracy of detected time of flight between the transmitted pulse and echo from the pipeline wall as well as the thickness of the pipeline wall was clearly improved. In order to testify the vality of the ultrasonic inspection device, inspection system was built in lab, the experimental results testified the efficiency of the ultrasonic inspection device and data anlysis software.All the approaches regarding ultrasonic signal feature extraction and information fusion can be used in other similar signal processing cases.