Studies On The Optimization And Regulations Of Gathering Network At The Chengdao Oilfield

There have been 10 years since the Chengdao, Shengli Oilfield was explored at 1993, and a reticular submarine pipeline and cable network have been built. According to the oilfield's gradual development, the production system was built by stages without a whole programming, as a result, the layout of pipeline network is irrational and flow distribution is uneven. Some production blocks are far from the central platforms and the pressure has reached 2.0MPa at the end of pipelines. High back-pressure is bad for oil lifting and safe operation of pipelines. According to the reservoir adjustment plan, 312 infill wells will be drilled in order to increase production capacity about 2.47 million tons, and No.3 Central Platform is also in planning. So it is necessary to optimize the structure of the existing pipeline network, and decide the connecting location of new wells as well as the diameter of new build pipelines.This paper researched the optimization of offshore oilfield gathering network. In order to optimize the layout of pipeline network, this problem was decomposed into three sub-problems: optimal production platforms grouping, optimization of central platform's location, and optimization of pipeline topology. Based on the concept of production-distance, a model was presented to optimize the grouping of platforms, the model was solved using genetic algorithm (GA). Optimal location of central platforms was decided by Warshall algorithm according to the weighted centre in graph theory. MST (minimum spanning tree) model and GA model were built to optimize layout of pipeline network. Prim algorithm was applied to solve the MST model. In order to resolve the GA model, an improved algorithm, single-parent genetic algorithm was proposed. Comparing the results of the two models, we find that: the network optimized using MST model has a minimum length of pipelines, however, its investment is not the minimum, and there are also some other problems, such as, the gathering distance is too long and the flow distribution is uneven. However, the GA model takes the investment of network as the optimization object. The network optimized using GA model has a lowest investment, and has a better hydraulic performance.According to the actual data of platforms at Chengdao Oilfield, the gathering network was optimized using the method presented in the paper. And the optimal network can save 15% of the investment compared with the existing pipeline network. In the paper, the location of No.3 Central Platform was optimized. The layouts of network before and after the central platform completed were presented respectively. Then the hydraulic performance of the network was simulated using PIPEPHASE, and the optimal diameters of pipelines new-built were calculated by the software. The simulation results show that the flow distribution in the network is uniform and the back-pressure of the platforms were controlled to around 1.0MPa.The study results can provide direct reference for the optimization of gathering network during the secondary development of Chengdao oilfield. And it also can be applied to the planning and design of gathering network for other offshore oilfield. And it will have a significant effect on improving the overall effectiveness of oilfield.