A Novel Particle Swarm Neural Network Model to Optimize Aircraft Fuel Consumption

Abstract

Due to the complex relations among the various factors, the nonlinear calculation of aircraft fuel consumption is very difficult. The purpose of this paper is to present a simplified method to estimate aircraft fuel consumption using a novel particle swarm neural network. Fuel consumption information obtained directly from QAR recorded flight data is trained by the neural network. The method can avoid the high cost of flight testing and wind tunnel testing. An improved particle swarm optimization algorithm embeds neural network topology to replace the network BP learning algorithm. The experimental results demonstrate that the proposed method integrates a new particle swarm neural network system, and significantly improves the system's learning ability and prediction of evolutionary effects.