Closed-Loop Control of Shock-Train Location in a Combusting Scramjet

This study demonstrates the closed-loop control of a shock-train location in the isolator of a Mach 2.2 direct-connect dual-mode scramjet facility (U.S. Air Force Research Laboratory research cell 18) operating on liquid fuel in ramjet mode. The shock-train location was established by the backpressu...

Full description

Saved in:
Bibliographic Details
Published in:Journal of propulsion and power 2018-05, Vol.34 (3), p.660-667
Main Authors: Vanstone, Leon, Hashemi, Kelley E, Lingren, Joe, Akella, Maruthi R, Clemens, Noel T, Donbar, Jeffrey, Gogineni, Sivaram
Format: Article
Language:English
Subjects:
Aerospace sciences
Algorithms
Control tasks
Identification methods
Liquid fuels
Mathematical analysis
Mathematical models
Predictions
Predictive control
Pressure ratio
Proportional derivative
Railroad accidents & safety
Supersonic combustion ramjet engines
System identification
Tracking errors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study demonstrates the closed-loop control of a shock-train location in the isolator of a Mach 2.2 direct-connect dual-mode scramjet facility (U.S. Air Force Research Laboratory research cell 18) operating on liquid fuel in ramjet mode. The shock-train location was established by the backpressure rise of combustion and actively controlled in a closed-loop fashion by actuating a water-cooled flap just downstream of the flameholder. The control task was to move the shock train to a user-selected location. The shock-train location was controlled using a proportional-derivative controller with real-time feedback of the shock-train location. Three algorithms for shock-train location were investigated: A threshold method that relied on wall-pressure measurements in the isolator, a predictive system identification-based mathematical model of the shock-train location and flap angle, and a predictive system identification-based mathematical model of the shock-train location and isolator pressure ratio. The first two methods were implemented experimentally within the closed-loop, whereas the third was examined through postprocessing. The threshold estimate closed-loop control scheme provided an approximately 3% (isolator length) tracking error; the predictive flap-position and pressure-ratio system identification predictions provided 10 and 7% tracking errors, respectively. All methods of shock-train location estimation could potentially facilitate closed-loop control of the shock train (individually or in parallel), with a tradeoff between the accuracy and information required for the implementation of each.
ISSN:0748-4658
1533-3876
DOI:10.2514/1.B36743