262
Y. Xu et al.
2.3 The Two-Stage Hybrid Algorithm
This Two-Stage Hybrid Algorithm consists of MGA and MPSO which all work well in
time-constrained problems because the optimal solution improves in every generation.
The MGA is first used for layout optimization of new pipelines and then MPSO is
employed to determine the design parameters, including the diameter of each pipeline,
the locations of new CSs and the optimal operation conditions of compressor facilities.
The process procedures of this two-stage hybrid algorithm are shown in Fig. 2.
Fig. 2. Topological structure of gathering system
However, the locations of new pipelines are hard to determine due to the complex
topology structure and numerous constraints. Most of the algorithms cannot give a
reasonable solution under limited iterations. To decrease the complexity of this problem
and obtain a better approximate solution, the proposed method reduces the number of
candidate pipeline layouts by some heuristic rules.
In this paper, we first simulate the pipeline network based on the field production
conditions and find certain nodes in which the gas flow cannot be transported to the
downstream stations. If we cannot find it then there is no need to build new pipelines,
otherwise, both of those unreasonable nodes and it is upstream nodes can be chosen
as the start point of the new pipelines. Then the candidate new pipelines can be determined under the limitation of pipe length, terrain, and obstacles. To further simplify this
Y. Xu et al.
2.3 The Two-Stage Hybrid Algorithm
This Two-Stage Hybrid Algorithm consists of MGA and MPSO which all work well in
time-constrained problems because the optimal solution improves in every generation.
The MGA is first used for layout optimization of new pipelines and then MPSO is
employed to determine the design parameters, including the diameter of each pipeline,
the locations of new CSs and the optimal operation conditions of compressor facilities.
The process procedures of this two-stage hybrid algorithm are shown in Fig. 2.
Fig. 2. Topological structure of gathering system
However, the locations of new pipelines are hard to determine due to the complex
topology structure and numerous constraints. Most of the algorithms cannot give a
reasonable solution under limited iterations. To decrease the complexity of this problem
and obtain a better approximate solution, the proposed method reduces the number of
candidate pipeline layouts by some heuristic rules.
In this paper, we first simulate the pipeline network based on the field production
conditions and find certain nodes in which the gas flow cannot be transported to the
downstream stations. If we cannot find it then there is no need to build new pipelines,
otherwise, both of those unreasonable nodes and it is upstream nodes can be chosen
as the start point of the new pipelines. Then the candidate new pipelines can be determined under the limitation of pipe length, terrain, and obstacles. To further simplify this