Optimum leak detection and calibration of pipe networks by inverse transient analysis

Inverse transient analysis is known as a powerful approach for leak detection and calibration of pipe networks. This work aims to introduce new features to carry out this analysis more efficiently. An optimization programming is developed to generate transients in the network optimally, and then pressure fluctuations are sampled at a few optimum measurement sites. Criteria are introduced for optimum site design. A non-linear programming with the least squares criterion is defined and then the unknown parameters of leaks and friction factors are inversely sought using sequential quadratic programming. A theoretical demonstrative example is also presented and solved using the method in which the obtained results show how the proposed new features make the process of leak detection more effective.