Skip to Main Content
Neural networks such as multilayer perceptrons and radial basis functions networks have been successful in a wide range of problems. A new modeling method is proposed for aircraft fuel pressurization ejector system based on support vector regression (SVR), a new class of kernel-based techniques introduced within statistical learning theory and structural risk minimization. This new modeling approach leads to solving convex optimization problems and also the model complexity follows from this solution. By using SVR with RBF kernel function, the SVR model of aircraft fuel pressurization ejector system is offline established. The simulation results show that the modeling precision is very high and the generalization capability of SVR model is also very good.