Pipeline Network Online Simulation Filter Under The Optimization Theory

With the boom in the production and application of oil and gas in China, the building of oil and gas pipeline is rapid increasing and a large-scale complicated pipeline transit system has been formed. In order to make the whole intricate process including the production, transit and supply of the oil-gas pipeline work well, the online simulation technology for pipeline network is more widely available. However, the process of pipeline network online real-time simulation is often interfered by the noise, which reduces the precision of online simulation result. With the view of acquiring the precision of online simulation effect, a study on pipeline network online simulation filtering method which can filter out the noise in online simulation process and enhance the precision of pipeline network online simulation is needed.This thesis studies a large amount of literature review at home and abroad, keeps up with the latest academic achievement in the field of nonlinear filtering technology and makes a deep study on the combination of the achievement with online simulation technology for pipeline network.With the study and research on the existing pipeline network simulation methods, this thesis builds up a fluid dynamics model under pipeline network online simulation by employing implicit method and makes a deep study on the noise inference occurring in the system of pipeline network online simulation, based on the established pipeline network online simulation model. Regard to the pipeline network state noise and measurement noise in the system, the thesis establishes objective function based on the state equation and measurement equation of pipeline network online simulation system, minimizes the established objective function of filter under the guidance of Optimization Theory, then Least Square Estimation and Extremum Principle will be employed in the process of addressing the objective function, which turns the issue of nonlinear filtering into nonlinear optimization so that an optimum pipeline online real-time simulator will be designed with the function of filtering the noise in the pipeline network online simulation system to meet the requirement of ensuring the precision of pipeline network online simulation result.The programs entitled Pipeline Network Online Simulation Filter under the Optimization Theory by employing Matlab 2010 a and Extended Kalman Filter are designed in this thesis, and a certain natural gas pipeline network system is being taken as an example to make filter calculation. The calculated data will be made comparison with noise-free data, noise-involved data and extended Kalman filter data, which can show that the studied pipeline network online simulation filtering method in the thesis has a fairly good effect and the precision of the filtering result is superior to the most-widely used Extended Kalman Filtering Method so that it is reliable to prove the applicability and validity of the studied filtering method in the pipeline network online simulation system.