|
|
    
|
Unsteady Propulsion System AnalysisNASA GRC Sectored-One-Dimensional Combustor CodeThe NASA GRC Sectored-One-Dimensional (S1D) Combustor Code is a time-Accurate, sectored-one-dimensional, single-species, reactive CFD code for simulation, analysis and control of gas-turbine combustion systems. In particular, the code is developed to investigate those combustor configurations which are susceptible to, or exhibit so-called thermo-acoustic instabilities. The code uses established numerical techniques for integrating the equations of motion, but contains a unique routine whereby one dimensional regions of differing cross sections are coupled at their common boundary. Thus, the code accommodates the abrupt area changes characteristic of modern low-emission combustors while maintaining a simple one-dimensional structure throughout the remainder of the combustor. Because it is a physics-based code or simulation, unstable behavior (if it occurs) is manifested as the code integrates in time. That is to say, the instable behavior is not brought about by any sort of external forcing, or internal sub-model. It arises quite naturally and correctly due to interactions between heat release and acoustic pressure waves. Lean-premix, low emission combustor designs appear to be particularly susceptible to the phenomenon of thermo-acoustic instability. This is due, among other things, to their lack of cooling air and their acoustically stiff geometry. One possible solution to the instability problem is active or feedback control. Here, a signal from the system (e.g. pressure) is used to detect incipient instabilities, and some form of actuation (such as fuel flow perturbation) is used to suppress and restabilize the system. Successful active control design however, is greatly enhanced by accurate modeling and simulation of the combustor of interest, and by straightforward implementation of simulated actuation and control strategies. The essential physical phenomena should be correctly captured. On the other hand, dynamic characterization of the system required for control design necessitates parametric analyses and multiple simulation runs. This, in turn, places the practical requirement of high speed on the simulation. If the instabilities are predominantly oriented along one spatial dimension, and if consideration is limited to premix type combustion, then it may be possible to satisfy the simulation objectives of numerical speed (i.e. model simplicity), and physical accuracy. Such restrictions apply to a large number of combustors. The NASA S1D code achieves the goals just described with the added, essential, capability of accounting for the large and abrupt changes in cross sectional area typically associated with premix combustors. Quasi-One Dimensional Reactive Code for Design and Analysis of Gasdynamic-Based Propulsion SystemsThe Quasi-One Dimensional Reactive Code for Design and Analysis of Gasdynamic-Based Propulsion Systems is a time-accurate, quasi-one dimensional, single-species, reactive CFD code for design and analysis of gasdynamic-based propulsion systems. These include, but are not limited to wave rotor based pressure exchange devices, detonative or deflagrative mode constant volume combustors, Pulse Detonation Engines, and combinations thereof. The code uses established numerical techniques for integrating the equations of motion, but also contains original sub-models to account for dominant and unavoidable loss mechanism. It possesses robust boundary condition sub-routines that allows for different modes of flow (inflow, outflow, sub or supersonic) in any of the envisioned ducts to which a gasdynamic device must be coupled. Thus, the correct flowfield is evolved rather than being imposed (which often leads to failures when numerically integrating). Additionally, the structure of the code is versatile, allowing for devices with stationary or rotating tubes or channels of arbitrary cross section. Because the code is quasi-one dimensional, it runs relatively quickly. This in turn allows for parametric analysis and design of conceptual devices and existing experimental rigs. However, because it is CFD based, with validated sub-models, the code output is realistic and vastly superior to simple thermodynamic analysis techniques often employed on gasdynamic devices. Gasdynamic based propulsion systems are inheritantly unsteady, or time dependant. The mathematical equations that govern their operation cannot be integrated or solved in closed form. Attempts to analyze such systems with traditional steady-state analysis typically results in performance predictions that are overly optimistic. Furthermore, steady state analysis conveys no useful design information. That is to say, a device that relies on precise timing of valves, sparks, wave transit times, etc., is impossible to build when time is neglected. Furthermore, it has been shown historically that even when an unsteady analysis is made (for example the method of characteristics, or thin wave techniques), if the dominant loss mechanism are not considered, experimental articles built using this analysis perform well below predictions. Thus there is a need for a relatively fast, yet reasonably accurate simulation capability that is generic enough to accommodate disparate gasdynaimc concepts. This code fulfills that need. AvailabilityBoth simulations are non-commercial, research codes written in FORTRAN. Source files and brief user-manuals are available for each by contacting the Technology Transfer and Partnership Office, or the author, Dan Paxson. Contact
|
|
responsible official: sanjay garg |
