Abstract
In this article, a nonlinear model predictive control algorithm for a micro-turboshaft engine is designed. The control effect is verified by a bench test. First, a micro-turboshaft engine test bench is built, and the open-loop control experiment was carried out on it. Based on experiment data, a linear parameter varying prediction model is established. Then, by online rolling optimization based on multistep output prediction, together with feedback correction, a nonlinear model predictive control algorithm is obtained. The influence of algorithm parameters on the control effect is studied, and reasonable prediction period M, control period N, and control coefficient R are designed. Finally, the application of nonlinear model predictive control in micro-turboshaft engine is verified by bench test. The results show that with the changing of pitch angle, nonlinear model predictive control algorithm has a faster adjustment speed and smaller overshoot, compared with the conventional cascade proportional–integral control with feedforward. It is proofed that nonlinear model predictive control can be applied to a real turboshaft engine and has a better control effect.
Highlights
Model predictive control (MPC) is a model-based closed-loop optimization control strategy, which has been used for more than 15 years in the industry as an effective means to deal with multivariable constrained control problems.[1]
nonlinear model predictive control (NMPC) control target of micro-turboshaft engine is to ensure that the power turbine speed np is constant, so the established linear parameter varying (LPV) prediction model should be a linear model of power turbine near the command speed
The feedback correction module can modify the prediction model by modifying linear interpolation output of the LPV model. It first calculates the error of command speed npr and micro-turboshaft engine np and uses PI term of the error to correct linear interpolation output value u0 of the LPV model, eliminating the steady-state error existing in NMPC control system
Summary
Model predictive control (MPC) is a model-based closed-loop optimization control strategy, which has been used for more than 15 years in the industry as an effective means to deal with multivariable constrained control problems.[1]. NMPC control target of micro-turboshaft engine is to ensure that the power turbine speed np is constant, so the established LPV prediction model should be a linear model of power turbine near the command speed. The LPV model established in this article can obtain steady-state point (x0, y0, u0) and state space model parameters A and B of micro-turboshaft engine according to u interpolation at each time.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
