Study On Leakage Quantification Method Of Water Distribution Network Using Wavelet Transform

Leakage is the universal phenomenon of urban underground water supply network, in addition, leakage rate of cities is on the rise far more than the requirements of water utility. Leakage rate of water supply network is the important indicator which can measure the situation of operation of water supp Ly network, however, it is difficult to calculate due to the complexity of pipe, growwing expansion speed and the restritions of management. There exist five methods of leakage quantification which are commonly used internationally, namely water balance method, minimum night flow, leakage predication method, blind source separation, Kalman filter method, the first three is only to estimate leakage but not to calculate the leakage degree before leakage control; two methods behind which have great success make use of signal processing method but are immature.If leakage can calculate, it will provide basis for leakage control and forecasting the better economic level of leakage, then we can simulate the potential benefit. Depending on the pressure and quantitative leakage of water supply network, we can fit the function relationship between them to guide regional pressure leakage control and to establish leakage management strategy. At the same time, leakage quantification can change present situation of leakage, economize water resource and improve water efficiency. Therefore, it has great practical significance to study the leakage quantification method of water supply network.This paper studies on leakage quantification using blind source separation and a new method—wavelet transform based on the inlet flow signal and pressure signal which acquisite from example network, laboratory network and practical engineering network. The main research contents of this paper include three aspects:1. Establish calculation examples of network using EPANET and establish experimental platform of leakage quantification in order to simulate different working statas and different number of leak in typical water supply scheme, then acquisite flow and pressure signal. This provides a basic of data.2. Establish a leakage quantitative model of blind source separation. For the problem of the amplitude of the blind source separation, the solution of the equation is put forward to solve the problem of the flow balance. Then put forward a new method—wavelet transform and wavelet transform leakage quantitative model. It selects optimum combination of wavelet and decomposition level using correlation coefficient for calculation examples of network and experimental platform of leakage. After that, program wavelet transform quantization procedure and analysis the data. Analyse applicability of theory, accuracy of calculation between two methods;3. Calculate leakage based on blind source separation and wavelet transform and fit wavelet simulation leakage, blind source separation leakage and the average pressure of water supply network relationship; Comparate the effect of leakage quantification and pressure leakage fitting.