Active Mitigation of Unstart in Scramjets
Abstract
Scramjet isolator operability and unstart are key phenomena aected by inlet Mach number, angle of attack, increased downstrea,m pressure, downstream pressure perturbations and/or combustor heat release uctuations, inlet ow distortion due to changes in the,angle of attack, and changes in the wall conditions such as boundary-layer separation (and reattachment) and roughness. During opera,tion, these eects may occur simultaneously and with mutual interaction, leading to a signicant level of complexity. To ensure robu,st operability during ight, it is imperative that unstart does not occur. Computational and experimental uid dynamics have been us,ed for operability assessment of scramjet ow-path; however, signicant research challenges remain with these methodologies includin,g the model-form uncertainties and numerical errors. Robust adaptive control algorithms can deal with model complexity and uncertain,ty as well as system nonlinearity, thereby, ensuring robust started operation under the uncertainties. A crucial research questi,on for the success of adaptive control is to determine the key modeling information needed to facilitate fast and reliable adaptatio,n to the actual scramjet dynamics and disturbances. The number of variable operating conditions and state variables that aect scram,jet operability and unstart is huge. Even though most CFD methods are plagued with numerical errors and model-form uncertainties, th,rute-force approach to cover the large variable space with CFD is cost prohibitive. To overcome this problem, we propose dimensional,ly and spatially adaptive sparse grid (DASGrid) method non-intrusively coupled with CFD to provide a computationally cost-eective m,ethodology for identifying the key modeling information to facilitate adaptive control to the actual dynamics of scramjet ow-path.,A second key research question is to determine eective sensors and actuators to facilitate unstart control, especially their placem,ent, type, and specications on authority and bandwidth. It is proposed to develop an integrated code for com-putational uid dynami,cs and active feedback control (CFD-C) for in-situ computational assessment of the control algorithm(s). The DASGrid code non-intrus,ively coupled with the CFD-C code can assess the performance of sensors and actuators in controlling the indicator,e scramjet ow-path. Two case-studies are proposed for validation of the CFD model and demonstration of the CFD-C code. These case s,tudies will leverage the experimental datasets previously obtained at NASA Langley and AFRL-WPAFB. Finally, to support the developme,nt of scramjet for defense applications, there is a critical need to develop a methodology for assessing the performance and robustn,ess of a given sensor/actuator conguration and control architecture. The inter-disciplinary research is proposed by Prof. Ragin,i Acharya (PI) from University of Ten-nessee Space Institute and Prof. Dennis Bernstein (co-PI) at University of Michigan Ann Arbo,r. Both Profs. Acharya and Bernstein have extensively discussed this topic during the past 6 months. Recognizing that challenges may, arise due to lack of a history of collaboration between them, they have planned to give both Prof. Bernstein and his graduate stude,nts access to Prof. Acharyas com-puting cluster at UTSI so that there is uniformity of software and hardware resources facilitating, the integration of CFD and control codes. Regular communications between the two faculty and their stu-dents, has been discussed an,d planned so that both the stated goals and deliverables can be achieved as planned.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 13, 2022
- Source ID
- N000142212457
Entities
People
- Ragini Acharya
Organizations
- Office of Naval Research
- United States Navy
- University of Tennessee