Friction Coefficient Inversion Calculation Based on Quasi - newton Method and Particle Swarm Optimization

Abstract

Establishing the pipeline network model of oilfield water injection under known partial pressure through fuzzy clustering method. In order to solve the inversion of pipe element friction coefficient in the case that the joint pressure of water injection pipe network is partly known, the simulation mathematical model of network node pressure and the optimal mathematical model of pipe element friction coefficient inversion are established respectively. By combining the simulation of nodal pressure by quasi-newton method and the inversion of friction coefficient by particle swarm optimization, the ideal pipe element friction coefficient of the pipe network with the partly unknown part of the node pressure can be inverted to achieve the partial accurate solution and partial estimation solution of the pipe element friction coefficient.