Research on aero-engine performance seeking control based on the NN-PSM on-board model

Abstract

In order to improve the real-time performance of performance seeking control (PSC), a neural network-propulsion system matrix(NN-PSM) on-board model is proposed and applied to PSC. First, based on NN-PSM, a large-envelope, multi-variable on-board adaptive model is established. The PSM is extracted through a small deviation linearization method. The neural network is used to map the relationship between the flight conditions, engine control parameters, and the engine performance parameters, and designed a Kalman filter to estimate engine health parameters in real-time. Then, four PSC mode of maximum thrust, minimum fuel consumption, minimum high-pressure turbine inlet temperature, and minimum infrared radiation intensity are designed using LP optimization algorithm as optimize algorithm. Finally, the simulation results show that NN-PSM has much higher precision than Compact Propulsion System Model (CPSM). The PSC simulations show that compared with the PSC based on the conventional CPSM, the proposed method has much better real-time performance and get better engine performance, such as more thrust, less specific fuel consumption, and less turbine inlet temperature.