Chaotic characters based oil pipeline leak detection method and system

Pipeline transportation is the main way of crude oil and product oil transportation, pipeline leak detection and positioning in time is very important to safe operation of the pipeline. The currently existing leak detection methods based on negative pressure wave are consider the interference existing in steady oil transportation state as random disturbance, which cause the difficult to detect tiny leak. But the existing studies have proved that when the oil transportation is steady, the pipeline the pressure shows the chaotic characteristics. In this paper, according to the chaotic characteristic in oil pipeline transportation, we design a oil pipeline leak detection methods, in which the algorithm adopts the combination of median and wavelet filter method, anomaly detection based on RBF neural network and fuzzy min-max classification method. At last we give the structure of pipeline leak detection system, in which the proposed leak detection method is the key method.