Year: 2008
Type: article
Abstract: Model predictive control is a strategy well-suited to handle the highly complex, nonlinear, uncertain, and constrained dynamics involved in aircraft engine control problems. However, it has thus far been infeasible to implement model predictive control in engine control applications, because of the combination of model complexity and the time allotted for the control update calculation. In this paper, amultiplexed implementation is proposed that dramatically reduces the computational burden of the quadratic programming optimization that must be solved online as part of themodel-predictive-control algorithm. Actuator updates are calculated sequentially and cyclically in a multiplexed implementation, as opposed to the simultaneous optimization taking place in conventional model predictive control. Theoretical aspects are discussedbased onanominalmodel, and actual computational savings are demonstrated using a realistic commercial engine model. I.
DOI: 10.2514/1.30591
Institutions: Cleveland State University, DEVCOM Army Research Laboratory
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2119422386