Time Delay Compensation for Hardware-in-the-loop Simulation of a Turbojet Engine Fuel Control Unit Using Neural NARX Smith Predictor

Autor: Mostafa Nasiri, Morteza Montazeri-Gh, Amin Salehi, Elham Bayati
Rok vydání: 2021
Předmět:
Zdroj: International Journal of Control, Automation and Systems. 19:3309-3317
ISSN: 2005-4092
1598-6446
DOI: 10.1007/s12555-019-0875-x
Popis: Hardware-in-the-loop (HIL) simulation is an effective technique that is used for development and testing of control systems while some of the control loop components are simulated in a proper environment and the other components are real hardware. In a conventional HIL simulation, the hardware is an electronic control unit which electronic control signals are communicated between the hardware and the software. But, HIL simulation of a mechanical component requires additional transfer systems to connect the software and hardware. The HIL simulation can achieve unstable behavior or inaccurate results due to unwanted time-delay dynamic of the transfer system. This paper presents the use of Smith predictor for time-delay compensation of transfer system in the HIL simulation of an electro-hydraulic fuel control unit (FCU) for a turbojet engine. A nonlinear auto regressive with exogenous input (NARX) neural network model is used for modeling and predicting the FCU behavior. The neural model is trained by Levenberg-Marquardt algorithm and the training and validation sets are generated using the amplitude modulated pseudo random binary sequence (APRBS). The consistency of the experimental real-time simulation and off-line simulation shows the applicability of the presented method for mitigating the effect of unwanted dynamic of the transfer system in the HIL simulation.
Databáze: OpenAIRE
Externí odkaz: