Varying-parameter modeling and extended Kalman filtering for reliable leak diagnosis under temperature variations

In this work the problem of robustness to temperature variations in pipeline leak detection and isolation is considered, extending preliminary attempts in this direction. The approach is restricted to pipelines short enough so that the spatial variations of the temperature can be neglected, and only time variations are addressed. The methodology is here based on a model taking into account the variations in all the parameters subject to temperature changes, like friction factor, water density, etc. A temperature measurement is then used to compensate for such variations in an Extended Kalman Filter used as a state observer, and real-time experimental tests are provided showing how an accurate leak identification is achieved when this compensation is done, while it fails otherwise.