| In the current petrochemical industry,the optimization of oil and gas pipeline network layout has been increasingly emphasized with the discovery of new oil and gas fields or the development of new blocks in older fields.As the main component of oilfield construction,the construction cost of oil and gas gathering and transportation systems accounts for a high proportion of the total investment in oilfield construction,and the results of pipeline layout directly affect the total investment in oil and gas field construction.However,there are many potential factors in pipeline layout environment that can cause pipeline failure,which may lead to pipeline leakage and more serious accidents,causing significant damage to personnel and property.Therefore,based on the actual situation of oil and gas fields,it is of great significance to adopt efficient intelligent optimization algorithms to solve the layout of oil and gas gathering and transportation networks and to construct an economical and safe pipeline layout.To address the diversity and poor convergence of existing multi-objective optimization algorithms,a multi-objective dual particle swarm optimization algorithm based on adaptive parameters(AP-MODPSO)is proposed.The algorithm divides the particle swarm hierarchy based on the difference in particle fitness,using Pareto ranking to divide the particle swarm into development group particles with better fitness and exploration group particles with poorer fitness,and the number of particles in the two groups adapts to the iteration rate.Then,the algorithm improves some parameters in the velocity updating formula of the original particle swarm algorithm,including the learning factor and inertia weight.The learning factor in the velocity updating formula of the two particle swarms is also adaptively changed according to the iteration rate,and the inertia weight is set using an evolutionary factor based on particle distance.AP-MODPSO algorithm is compared with four other multi-objective particle swarm optimization(MOPSO)algorithms through simulation experiments on three series of multi-objective test functions.The results show that the AP-MODPSO algorithm has better diversity and convergence in solving similar multi-objective models.For the multi-objective optimization problem of oil and gas pipeline network layout,a multi-objective optimization model of oil and gas pipeline network layout is constructed to minimize the construction cost and potential accident risk of the pipeline network.The layout plan satisfying different objectives is realized by solving the model.Firstly,by comparing the common topology structures of oil and gas pipeline networks,a two-level tree topology structure with the advantage of saving pipeline length is selected.Secondly,considering the obstacles in the model,the cluster division is initially determined based on the distance between wells and stations and the possible detour distance.Then,the pipeline connection method between each cluster of wells and stations and between each level of gathering stations is determined by using the Delaunay triangulation algorithm and the Prim algorithm,and the pipeline diameter specifications are determined based on the constraints of oil and gas flow rate and economic flow rate.Finally,the construction cost and potential accident risk of the pipeline network can be calculated according to the objective function,and the model can be solved by the AP-MODPSO algorithm to obtain multiple layout plans that satisfy different objectives.Through simulation experiments on an oilfield with 67 wells and stations,the effectiveness and feasibility of the proposed multi-objective optimization model and AP-MODPSO algorithm for oil and gas gathering and transportation pipeline network layout are verified. |
