Denoising of Direct Simulation Monte Carlo Data Using Proper Orthogonal Decomposition Technique

Sie sind hier: Homepage > Suche

Denoising of Direct Simulation Monte Carlo Data Using Proper Orthogonal Decomposition Technique

Kumar, Rakesh / Chinnappan, Arun K. / Akhil, M. V.

The direct simulation Monte Carlo method is one of the most widely used particle-based methods in solving flows with a large degree of rarefaction. Direct simulation Monte Carlo method finds its place in effectively simulating flows in rarefied regimes like the high-altitude flows and micro/nanoflows. The method, although effective, is prone to statistical fluctuations or noises. To bring out the true signal from the direct simulation Monte Carlo simulation, a large amount of sampling data is required, especially at low speeds. This demands higher computational power and related expenses. In this work, a novel postprocessing technique based on proper orthogonal decomposition is proposed to reduce statistical scattering associated with direct simulation Monte Carlo simulations. The mathematical technique of proper orthogonal decomposition as applied to the data obtained from the particle-based direct simulation Monte Carlo method is discussed in detail. The postprocessing technique is shown to be very effective in reducing the computational time required to achieve converged low-noise direct simulation Monte Carlo results, without incurring any significant computational expense.