چكيده به لاتين
In gas pipeline networks, the set-points should be carefully tuned to minimize the fuel consumption of compressor stations and meet the network requirements. In practice, the real demand has some variations over the forecasted one. Consequently, utilizing an appropriate controller to minimize the fuel consumption and manage the network variations is inevitable. The model predictive control is a great choice for systems with long delay such as gas networks. In this thesis, a novel approach based on intelligent algorithms and three basic functions is proposed for modeling the transient state of gas pipeline networks. Using the proposed technique, each pipeleg is analyzed separately which speeds up the analysis. Also, two control strategies for the nonlinear model predictive control of a gas pipeline plant are proposed. The control of the network is carried out in two open loop and closed loop steps. In the open loop step, the fuel consumption of the compressor stations is optimized considering the known demand of the network. Then the output pressures are considered as the reference trajectories in the closed loop step. A novel approach is proposed based on metaheuristic algorithms for optimization of a gas pipeline network in fully transient conditions. The optimization is carried out by a straightforward methodology in each time sample, which leads to more precise solutions as compared to the quasi transient optimizations. In the first approach, a multi-layer perceptron neural network is used to model the gas pipeline network in transient state as the plant and an intelligent optimization algorithm is used to minimize the control error. In the second approach, instead of using an intelligent optimizer, a neural network is used to control the plant. The prediction power of the neural network is used to predict the plant output over a receding horizon. Initially, the neural network is trained offline and its weights and biases are used as the initial values in the online training. The proposed strategy consists of two main stages. In the first stage, the compressor set-points are optimized in the open loop condition considering the forecasted demand over a receding horizon and the resulting output pressures are chosen as the reference trajectories for the closed loop system. In the second stage, the controller is applied to compensate the demand variations. Numerical results confirm the accuracy and robustness of the proposed controller in the presence of demand variations, noise and uncertainties.
