A pre-filter combined with threshold self-learning wavelet algorithm is proposed for hydraulic pressure signals denoising. The denoising threshold is self-learnt in the steady flow state, and then modified under a given limit to make the mean square errors between reconstruction signals and desirable outputs minimum, so the corresponding optimal denoising threshold in a single operating case can be obtained. These optimal thresholds are used for the whole signal denoising and are different in various cases. Simulation results and comparative studies show that the present approach has an obvious effect of noise suppression and is superior to those of traditional wavelet algorithms and back-propagation neural networks. It also provides the precise data for the next step of pipeline leak detection using transient technique.
A further development of exclusively inverse frequency domain method for leak detection in pipelines is presented and validated.The location and leakage can be determined by analyzing the difference of transient water head response between the simulated and measured data in frequency domain.The transient signals are generated by portion sharp closure of a valve from the small constant opening and it needs only a few meters of water.The discrete boundary conditions and observation data are both transformed in frequency domain by Laplace transform.Example in numerical simulation is studied for demonstration of this approach.The application of the method to an experimental pipeline confirms the analysis and illustrates successful detection of the single pipeline leak.The precalibration approach is presented to minimize the effect of data and model error and it splits the method into two parts.One uses data from a known state to fit the parameters of the model and the other uses data from the current state for the fitting of leak parameters using the now calibrated model.Some important practical parameters such as wave speed,friction in steady and unsteady state and the adaptability of the method are discussed.It was found that the nonlinearity errors associated with valve boundary condition could be prevented by consideration of the induced flow perturbation curve shape.
The leak detection is of great importance in the reliable operation and management of a pipeline system. Recently, attention is shifted to the use of the time domain or frequency domain methods based on the transient analysis. These methods sometimes require accurate pressure signals obtained during the transient period or by creating ideal conditions in testing. This paper proposes a method that does not require transient simulations over the whole or an extended period of time, but uses the first transient pressure oscillation to detect leaks. The method considers the propagation of the pressure oscillation wave created from a fast valve closure and the reflected damp wave from the leak. A leak in the pipe gives rise to reflected waves which in turn create discontinuities in the observed signal at the measurement section. The timing of the reflected damp wave and the magnitude represent the location and the size of the leak, respectively. An analytical expression is derived based on the Method Of Characteristic （MOC） for the relationship between the leakage and the reflected magnitude. The leak detection procedure based on the method is also given. Then the reliability of the method is tested on numerically simulated pressure signals and experimental pressure signals with calibrated leak parameters, and the results indicate a successful application and the promising features of the method.
