Pipe Leak Detection And Location Method Based On Chaotic Characteristic And Recurrence Plot

In recent years, oil and gas production has been more and more important to our life. However, with the development of the cause of China’s pipeline, As part of the pipeline to use time longer and longer, Construction process of the defects and natural corrosion caused oil leakage of oil conveying pipeline, Especially, the man-made destruction and the action of bunkering has brought great threat to the safety of the people, the property of the state, the natural environment. So when the oil pipeline leakage, can timely discover the leakage and locate the leak location becomes more and more important.Paper describes the research status of pipeline leak detection and location, analyses the commonly used oil pipeline leak detection and location method. For single parameter research pipeline leak can lead to false alarm or positioning were some drawbacks, This article is based on system of Supervisory Control And Data Acquisition(SCADA), for processing flow and pressure double parameters, introducing chaos characteristic analysis and recurrence plot method for pipeline leak detection and location method are analyzed. The main contents include:(1) Analyzed the leakage of oil conveying pipeline flow and pressure changes in the situation, put forward for processing flow and pressure double parameters, chaotic characteristic analysis and recursive diagram method was adopted to realize pipeline leak detection and location. Negative pressure wave locating formula are given, and negative pressure wave speed is improved in the implementation process of orientation.(2) Taking the pipeline entrance flow signal as the processing object, based on chaotic characteristic analysis to identify pipeline leakage. First,on the system dynamic characteristic of phase space reconstruction, and in order to avoid the inconsistency of calculating separately,using the differential entropy method to calculate the delay time and embedding dimension at the same time. then uses the Wolf algorithm to calculate the index, and determine the characteristics of chaos system, and finally by calculating correlation dimension of GP algorithm analysis chaos characteristic of the system.(3) Using the dual tree complex wavelet transform for processing the signal that was collected by SCADA system, feature extraction, judge whether the leakage of oil pipeline, then using the recurrence plots and average gray method for characteristic point positioning, determine the characteristics of the time, According to the other end determined characteristic time can determine the negative pressure wave transmission to the ends of the time difference. The final combined positioning formula and negative pressure wave velocity determine the leak location.