Research Of The Leakage Detection Based On Negative Pressure Wave Method

With the unique advantage of transmitting gas, liquid and slurry, pipeline has become the fifth important conveyance after railway, highway, waterway and voyage. Because of inevitably pipeline aging, erodibility, stiletto steal and so on, leakage accident happened frequently. It interferes in natural production badly, brings enormous economic losing and pollutes the environment. So the research and application of the leak detecting and locating system of transporting-oil pipeline becomes an urgent problem. Advanced research of leakage detection and location of pipeline can find the leakage of pipeline and take action quickly. It may reduce the happening of steal case, avoid the pollution of environment, reduce the interprise's economic losing and improve the level of interprise's automatic management.In this, a variety of domestic and leak monitoring techniques have been studied and compared in light of reality, to improve the negative pressure wave propagation velocity formula for the elimination of the first, the end of the dynamic response time of pressure sensors bad way to unify the first use of GPS data collection terminal system, system time, to improve positioning accuracy. Analysis of the leakage signal characteristics of signal extraction and processing of computer-related experiments, the application for the site filtering denoising method of wavelet transform. Artificial neural network method applied to the pipeline to identify the characteristics of stress on the signal, the extracted feature vector can be used as the basis for leak detection, to determine whether the leak happened. At the same time take into account the leak to determine the accuracy of the RBF neural network to identify the operational status of piping may be based on a large number of extracted feature vectors to determine the occurrence of leakage. MATLAB and LabVIEW combines the advantages of the implementation of the interface between the two.