Skip to Main Content
Neural networks (NNs) have been successfully used for implementing control architectures for different applications. In this paper, we examine NN augmented intelligent control of a turbo-fan engine toward the goal of minimizing a performance measure on-line. This architecture utilizes an adaptive critic to estimate the engine performance, which is then used to train an NN demand generator for minimizing the performance measure. The present architecture is implemented on a nonlinear model that was provided by General Electric. The model simulates a changed engine by changing the flow and efficiency scalars of the various components of the engine. Results of using the adaptive critic-based performance seeking control architecture show excellent improvement in performance over time.