关键词: 集输管网, 湿天然气, 高含硫, 遗传算法, 全局优化, 阀门数目

Abstract:

In view of characteristics of the natural gas from the Puguang gas field, a mathematical model for the global optimization of tree-type network layout, pipe diameters and valve numbers, was established, which had an objective to minimize the investment cost of a gathering pipe network construction and took the flow rate, pressure and pipe diameter of each nodal point of the network as constraint conditions. A dual coding genetic algorithm that combines binary coding with integer coding was adopted to solve problems encountered in optimizing tree-type pipe network layout, pipe diameters and valve numbers simultaneously in order to minimize the total investment cost. Given the gas-condensate flow in pipelines, models for calculating physical properties and temperature drop, the wet gas from the Puguang gas field were established respectively, and a hydraulic calculation model which is applicable to miscible delivery pipelines and capable of high calculation accuracy was chosen by comparing the calculated data with measured ones, which ensured the check veracity for calculations of physical properties, pressures and temperatures during the optimization and for the constraint condition of optimized models as well. Practical examples have verified the accuracy of the proposed optimization model and the feasibility of the optimized algorithm. This approach is of guiding significance both in reducing the investment cost of a pipeline network construction of the acid gas and in enhancing the security of gathering pipe networks.

Key words: gathering pipe network, wet gas, high H₂S content, genetic algorithm, overall optimization, number of valves