Problemas inversos em escoamentos transientes de líquidos em tubulações

Abstract

Transient flows in pipes is a topic that has been widely studied both within the industrial and academic context. In hydraulic systems, the transient signals in the flow usually arise due to the way in which elements present in the pipeline, such as hydraulic pumps and valves, are operated. If the speed of the fluid is changed abruptly, a phenomenon called water hammer or hydraulic transient can occur. This phenomenon is characterized by the appearance of pressure waves of great magnitude and speed, which can be very dangerous for the integrity of the entire hydraulic system. Thus, in the design of a hydraulic system, it is very important to characterize how this phenomenon occurs and its possible consequences. Thus, the present work seeks the application of Bayesian techniques for the solution of an inverse problem in order to estimate parameters present in the mathematical formulation of water hammer. For that, the Markov Chain Monte Carlo method will be applied in the classic way and using parallel computing in the hydraulic transient phenomenon simulated in an experimental hydraulic circuit present at the Technology and Heat Transmission Laboratory (LTTC / UFRJ) in the final stage of construction.