Font Size: a A A

Study On Overall Optimization Design Of Gas Field Star-Tree Shape Pipeline Network Layout In Discrete Space

Posted on:2020-05-17Degree:MasterType:Thesis
Country:ChinaCandidate:S Q JingFull Text:PDF
GTID:2381330602958326Subject:Oil and Natural Gas Engineering
Abstract/Summary:PDF Full Text Request
In the construction process of natural gas field ground engineering,the gathering pipeline network system is the core of the ground engineering,and the investment cost is huge.At present,most of the methods for solving the layout design of gas pipelines in natural gas fields are hierarchy optimization and progressive optimization.The gas field is not considered as a whole,and the results may not meet the gas transmission requirements.In this paper,the overall optimization design of the distribution of natural gas field star-tree shaped pipeline network based on discrete space is carried out.The hierarchy optimization of gas field star-tree shaped pipeline network layout in discrete space is studied:the characteristics of discrete space are studied,the structure of natural gas pipeline network is introduced,and the layout structure of star-tree shaped pipeline network is described in detail and the corresponding mathematical model and design idea of hierarchical optimization are proposed.Based on the idea of the algorithm,the overall optimization design of gas field star-shaped pipe network layout is studied.The overall optimization algorithm of gas field star-shaped pipe network layout based on discrete space is proposed,which optimizes the layout of gas field star-shaped pipe network.This paper mainly completed the following research results and conclusions:(1)Considering the characteristics of discrete space,the structure of natural gas pipeline network is introduced.The layout structure of star-dendritic pipe network is described in detail,and the corresponding mathematical model and hierarchical optimization design ideas are proposed.The optimization of the middle well group division,the station/valve site selection,the dry branch pipe network connection and the pipe network structure parameters are optimized for the well group-valve site selection,the dry branch pipe network connection optimization and the pipe network structure parameters optimization.(2)Introduce the characteristics of traditional genetic algorithm,and take the genetic algorithm as the idea,consider the structural characteristics of the gas field star-shaped branching network,and design the algorithm,and design the optimization algorithm of the gas field star-shaped pipe network classification.The realization includes the well-valve corresponding overall coding,the development of the pipe network solution adaptive function,the design improvement based on the coding design,the selection method,the crossover and the mutation operator,and the improvement in the OR-LIBRARY database.The test data is compared with other algorithms to verify the feasibility of the algorithm.The branching network connection of the star-shaped well group is designed.The calculation complexity is compared with the Kruskal for the branch network connection design.After the completion,the optimization of the pipe network structure parameters is discussed.(3)Based on the design of grading optimization of gas field star-shaped pipe network,the overall optimization design of gas field star-shaped pipe network layout based on discrete space is carried out.The well group-valve selection is optimized,the dry branch network connection is optimized and integrated into the star branch pipe network design optimization,and the star-shaped gas production pipe network solution and the branch gas collecting pipe network solution in the hierarchical optimization design are integrated into one solution.Ensure that the individual in each solution is a complete gas field layout design.(4)Based on the discrete space-based gas field star-shaped overall layout optimization model and the discrete space-based gas field star branch network layout optimization algorithm,the algorithm design of the overall optimization design method is studied.The normalization and covariance matrices are introduced.In the evolutionary solution of genetic algorithm,the consistent non-gradient and randomness of genetic algorithm are avoided,which indirectly increases the gradient decreasing trend of genetic algorithm and increases the evolution accuracy of genetic algorithm.In the mutation operation,the idea of "niche" is introduced,and the optimal solution is selected by comparing multiple "local inheritance" codes selected by the mutation to achieve the result of improving the mutation efficiency;integrating the dendritic connections into the genetic algorithm In the individual solution method,the star shape and the branch shape are integrated into a set of codes to ensure that all individuals in the population in the calculation of the genetic algorithm are design schemes that satisfy the constraint conditions,and the 2-opts algorithm is used in the branch design.Combined with the kruskal algorithm,when the pipe network connection is made,combined with the pipe network structure parameter design,the solution consideration is more sufficient,and the integrity of the solution algorithm is further strengthened to ensure that the connection does not exist to satisfy the local star-shaped well group division.When it is better,the design of the branch line connection method cannot be obtained.(5)According to the overall optimization algorithm of the gas field star-shaped gathering pipeline network,the program compilation design,considering the complexity and readability of the program design,and the incompleteness of the layout of the gas field gathering pipeline network in this paper,The compiled program should have better scalability.According to the above situation,the java language is used,and the springboot+vue+mybatis framework is used to design the overall optimization algorithm of the star-shaped pipe network layout to ensure the program has good readability and scalability.(6)Taking a coalbed methane field in China as an example,based on the original pipe network layout design research,the classification optimization and overall optimization are used to solve the problem,and the design results of classification optimization and overall optimization are obtained and compared to verify the feasibility of the algorithm.According to the research,for the layout design of the gas field star-shaped pipe network in discrete space,the design-specific algorithm has an optimization effect on the model solution,and the overall optimization effect is better than the grading optimization.
Keywords/Search Tags:Pipeline network optimization, Improved genetic algorithm, Overall optimization, Layout optimization, Star-tree network
PDF Full Text Request
Related items