Study On Optimization For Oilfield Water Injection Pipeline Network Based On Particle Swarm Optimization

The water injection pipeline network is a large-scale network system, only the pipeline investment approximately composes about forty percent of the system's total investment. Optimizing the pipeline network can save the pipeline and massive funds. At present, the technology on the optimization of water injection pipeline network isn't mature enough, and the references can used are very few, we usually get something from the oil gas collection or the urban water supply and other pipeline network. Here we attempt to use particle swarm optimization (PSO) algorithm to optimize the network of the water injection, and try to obtain the better topology of the network, and get the shortest pipeline.PSO algorithm is a kind of evolution computing technique based on swarm intelligence algorithms. Comparing to other evolutionary algorithms, its convergent speed is fast, rules are simple, and it is easy to implement programming. In order to restrain particles from trapping in local optimum, here, according on the traditional mutation PSO and unifying the study factor, we propose the CVMPSO(Velocity Mutation Particle Swarm Optimization based on Coefficients)which leading into mutation operation on speed. The improved algorithm enables the swarm search fast in the initial iterate period and precisely search in the later period, speeds up the convergence rate, at the same time, it carries out the mutation operation for speed every iteration, so the algorithm can escape from the local maximum, raise the accuracy of non-linear optimization and keeps PSO algorithm structure simple in the meantime.The paper first describes and analyzes the network of the water injection, then establishes the mathematical models of pipeline network. Aiming at the complexity of the mathematical model, this article divides the question of layout optimization of water injection system into two sub-questions: First question, the partition optimize to the water injection well, so is best division of the water-injection well number, Second question, choose the position of water distribution station and water injection station and main stem. Then we carry out the optimization of the integer layout. In this article every mathematical model of the sub-questions is established, and each of which has its corresponding proper solution. Thus the main topology optimization to the layout of water injection system is settled effectively.This article mainly uses the PSO optimization algorithm which is based on the dynamic alignment study factor and carries out mutation operation on speed to design and optimize the layout of water injection network., and compares with the traditional position mutation PSO algorithm and the artificial project, finally the result indicate that this method is effective, comparing with the other two algorithms, it has obtained a shorter pipeline, saves the pipeline investment. The network of oil field system can be used for reference mutually, so, this paper also has some help to the other oil field system on some degree.