Uncertainty Quantification in Conceptual Design via an Advanced Monte Carlo Method

Abstract

A method for quantifying uncertainty in conceptual-level design via a computationallyefficient probabilistic method is described. As an example application, the investigated method is applied to estimating the propellant mass required by a spacecraft to perform attitude control. The variables of the design are first classified and assigned appropriate probability density functions. To characterize the attitude control system a slightly-modified version of Subset Simulation, an efficient simulation technique originally developed for reliability analysis of civil engineering structures, is used. The proposal distribution aspect of Subset Simulation is modified vis-à-vis the original technique to account for the general characteristics of the variables involved in conceptual-level design. The results of Subset Simulation are compared with traditional Monte Carlo simulation. The investigated method allows uncertainty in the propellant required to be quantified based on the risk tolerance of the decision maker. For the attitude control example presented, Subset Simulation successfully replicated Monte Carlo simulation results yet required significantly less computational effort, in particular for risk-averse decision makers. Nomenclaturelognormal random variable with parameters μ and σ m number of Subset Simulation levels N number of Subset Simulation samples N c number of Markov chains developed at each simulation level; N c = p 0 N

Extracting Rare Failure Events in Composite System Reliability Evaluation Via Subset Simulation

“…Subset simulation was originally developed for seismic risk analysis of building structures subjected to stochastic earthquake motions [20], [23], where the problem involved a large number (theoretically infinite) number of random variables arising primarily from the time-domain stochastic description of ground motions. Applications of SS to different disciplines have appeared, e.g., in aerospace engineering [24], [25], fire engineering [26], geotechnical engineering [27], [28], nuclear engineering [29], [30], and meteorology [31]. Performance of subset simulation in a set of benchmark problems is presented in [32] and [33].…”

A collocation interval analysis method for interval structural parameters and stochastic excitation

“…In structural static and dynamical analysis, the structural parameters are usually subject to variation due to fluctuations in material properties, uncertainty in boundary conditions, and variations caused by manufacturing and assembly techniques. In the current literature regarding structural response problems with random uncertainties, there are three main methods [2]: the Monte Carlo simulation method [3][4][5][6], the stochastic finite element method [7][8][9][10], and the orthogonal series expansion method [11,12]. The Monte Carlo simulation method is very efficient in the aspect of structural random analysis, but it is quite time-consuming.…”