Abstract: When an oil oilfield enters into a high water-cut stage, the metering station in gathering system is often inefficiently utilized. In this background, considering the situations of reduced metering stations, recycling of existing pipelines and construction of new pipelines, a topological optimized mathematical model for gathering system is built, with the minimum total cost as the objective function and with well/station ownership, metering capacity, geometric position, pressure limit and pipe diameter as constraints. Since this model is nonlinear, well/station ownership is solved and optimized by improved hybrid genetic simulated annealing algorithm. Based on the physical significance of objective function and constraints, the constraints can be processed to facilitate the realization of the algorithm. The solution efficiency of algorithm is increased by improving the crossover and mutation operators. The optimization results in actual application provide the specifications and recovery expenses of new pipelines, proving the effectiveness of the optimization model and algorithm.