Transient Simulation of Natural Gas Network by Hybrid Taguchi Binary Genetic Algorithm

摘要

The analysis of gas transportation networks is the backbone for further processes such as optimization and control. The static analysis is based on algebraic equations which are straightforward and easy to solve, but may result in solutions far from the optimum due to the dynamic nature of the network. Hence, the transient analysis is inevitable. It is based on a set of equations containing partial differential equations (PDEs) for each pipeleg (Navier-Stokes equations), algebraic equations of compressors, the initial conditions and the boundary values. Since the governing equations of each pipeleg are PDEs, the internal boundary values of the network should be considered according to the topography of the pipelegs in the network, which makes the traditional transient analysis complicated and time consuming. In this paper, a straightforward method based on metaheuristic algorithms is proposed for the transient analysis. Using the proposed technique, each pipeleg is analyzed separately which speeds up the analysis. The source flow rates are considered as the optimization variables and based on them, the demand pressures are calculated. The sum of the absolute differences between the real demand pressures (known as the boundary values) and the calculated ones is the error of the proposed modeling. To minimize the error, a powerful metaheuristic algorithm called Hybrid Taguchi Binary Genetic Algorithm is utilized. Numerical results confirm the efficiency and accuracy of the proposed method that leads to near-zero error.