Estimação de estado em malhas de escoamento de gás natural: uma proposta para malhas complexas
Natural gas is economically important due to its low value and less environmental impact. Brazilian gas is produced mainly from offshore fields in deep water, where pipeline networks are in operation in order to transport this natural gas to onshore processing units. As a decision support tool, flow models integrated with field measurements are used, which estimate variables at different points in the flow. This work seeks to estimate the state in an offshore gas pipeline network in deep
... including the estimation of flow parameters, to improve the prediction of pressure and temperature profiles and greater tolerance to inaccurate measurements. To achieve these objectives, the Kalman filter is used. A single-phase, transient, thermal and compositional flow model is proposed for the reproduction of flow dynamics in a gas pipeline network; this model is used in the extended Kalman filter to describe the dynamics of the system. Available cases in the literature are discussed, as proposed onshore and offshore gas pipelines, a simulated offshore network and a real network of pipelines in operation, in order to verify the scenarios where this proposal presents gains. The results show that the incorporation of the energy conservation equation in the propagation model on offshore pipelines improves the accuracy of profiles and flow parameters simultaneously estimation, for pressure and temperature profiles in scenarios with deep water flow transients. In addition, the approach was validated with field data for an offshore network in operation, presenting consistent results for the estimated parameters, as well as robustness to the non-representative measurements on the sensors. These are the main contributions of the Thesis.