Optimization of Multistage Stations Locating in Oil Transportation System Based on a Hopfield Neural Network Simulation Machine

A large-scale nonlinear MIP model of optimum locating of multistage stations in oil transportation is established. Because the model is very difficult to solve by traditional methods, a synthetic solution is presented by a Hopfield neural network algorithm. In establishing the optimization model, the real continuous variables are changed into discrete 0-1 integer variables so that the nonlinear MIP model is transferred into a pure 0-1 nonlinear IP model and it is possible to solve the model with high speed because the whole solving process falls into a binary calculation environment. In the solution, a simulating machine based on the Hopfield neural network model is developed by C++ computer language, the structural parameters of the machine are deduced from the optimization model. A practical application shows that the simulating machine can find optimization results with high speed