A size-transferring radial basis function network for aero-engine thrust estimation

摘要

Thrust regulation plays an important role in the aero-engine control. However, the thrust is unmeasurable in flight which poses a great challenge to the thrust control. Traditional thrust control methods are implemented by controlling the parameters tightly related to thrust and reserve enough safety margins to protect the engine. To realize the direct thrust control, the methods to estimate thrust is urgently required. In this paper, a new algorithm based on particle swarm optimization (PSO) and radial basis function neural network (RBFNN) is proposed to estimate the thrust. A strategy named "size-transferring" is developed to select and adjust the network size of the RBFNN. Besides, to solve the high-dimensional optimization problem during the estimation, a new approach based on the PSO algorithm is also illustrated. The successful application of the proposed algorithm to the aero-engine thrust estimation problem demonstrates its effectiveness.