Sampling based on Sobol' sequences for Monte Carlo techniques applied to building simulations

Monte Carlo (MC) techniques are commonly used to perform uncertainty and sensitivity analyses. A key element of MC methods is the sampling of input parameters for the simulation, where the goal is to explore the entire input space with a reasonable sample size (N ). The sample size determines the computational cost of the analysis since N is equal to the required number of simulation runs. Quasi-random (QR) sequences such as the Sobol′ sequences are designed to generate a sample that is uniformly distributed over the unit hypercube. In this paper, sampling based on Sobol′ sequences is compared with other standard sampling procedures with respect to typical building simulation applications. The work revealed that for the most of the analyzed aspects the sampling based on Sobol′ sequences performs better than the other investigated sampling techniques.