Abstract

This paper proposes a technique for the preliminary identification of pipes obstructed by blockages within a pipeline or a complex pipe network, relying only upon few measurements that are usually gathered to monitor the infrastructure operability. Blockages are identified as a result of an optimization procedure that minimizes the discrepancy between measured and simulated flow quantities. The framework exploits a Finite Element-like simulator of the flow problem in the system of pipes and Genetic Algorithms to perform the optimization procedure. Thermal properties of the fluid flowing inside the pipe system are considered. Results of the identification procedure are presented, and advantages and disadvantages of the procedure for the on-field implementation are outlined.