A Reduced Order Lunar Lander Model for Rapid Architecture Analysis

Thumbnail Image
Thompson, Robert W.
Krevor, Zachary
Young, David A.
Wilhite, Alan
Associated Organization(s)
Supplementary to
The conceptual design of an architecture for space exploration involves the evaluation of many concepts. These design spaces may encompass millions or billions of options when each trade is evaluated at the system, vehicle, subsystem, and component level. Various techniques are typically employed to select the configuration of systems that best meets the requirements of the architecture. These include multi-attribute decision making techniques as well as optimization with the use of genetic algorithms and other stochastic methods. In order to speed up the evaluation of these options, a set of reduced-order vehicle models can be used. These models evaluate the gross weight, dry weight, cost, and reliability of a vehicle given a set of programmatic and performance options in less than a second, versus the use of design codes that take on the order of minutes to hours to converge to a vehicle design. The use of such reduced-order models also enables other techniques that would otherwise take too long to run, such as Monte Carlo simulation to model uncertainty, as well as optimization of the vehicle and studies of sensitivities to changes in programmatic and performance inputs. A reduced-order lunar lander model is presented, utilizing response surface equations (RSEs) in place of detailed disciplinary simulations. While some fidelity is lost in approximating these disciplines with RSEs, this approach can be used to evaluate the relative impact of various trade studies at the subsystem, vehicle, and architecture levels. The propulsion system is modeled using a response surface of the REDTOP-2 code. In a similar manner, the trajectory for lunar descent and ascent is simulated using Program to Optimize Simulated Trajectories (POST), and then approximated with a RSE for use in the reduced-order lunar lander model. The weights and sizing model of the lunar lander is based on a combination of historical mass estimating relationships (MERs), and physics based mass estimating relationships. Development and production cost modeling is performed using the Cost Estimating Relationships (CERs) from the NASA-Air Force Cost Model (NAFCOM). Because the reduced-order lunar lander model evaluates rapidly, stochastic optimization methods such as genetic algorithms can be used to find the performance inputs (such as thrust-to-weight ratios, propellant choices, and expansion ratios) that optimize the vehicle for smallest mass, highest reliability, or smallest development cost. A user-customizable Overall Evaluation Criterion (OEC) can be used to optimize the vehicle for a weighted combination of multiple criteria. Within an architecture analysis, this quick turn-around is useful for rapidly designing the lunar lander to meet the mass constraints of the launch vehicles, and the cost and reliability constraints of the programmatics.
Date Issued
Resource Type
Resource Subtype
Rights Statement
Unless otherwise noted, all materials are protected under U.S. Copyright Law and all rights are reserved