Paper detail

Support Vector Spectrum Approximations: Efficient Calculations of Light Scattering Lineshapes

Measuring scattered light is central to many laser-based gas diagnostic techniques, e.g., coherent anti-Stokes Raman spectroscopy (CARS) and filtered Rayleigh scattering (FRS). To produce quantitative measurements with such techniques, a computational model of the scattered spectral lineshape is necessary. While accurate, these models are often quite computationally demanding and thus cannot be used in situations where computational speed matters. To overcome this, approximations of these spectral lineshape models can be used instead. In this paper, we develop a method called support vector spectrum approximation (SVSA). This method uses machine learning to create efficient and accurate approximations of any existing spectral lineshape model. The SVSA framework improves upon existing methods by allowing efficient approximations of spectral lineshapes to be calculated in arbitrary flow regimes. We demonstrate the efficacy of SVSA in approximating coherent and spontaneous Rayleigh-Brillioun spectra. We also show that SVSA reduces the computational cost of a simulated filtered Rayleigh scattering experiment by a factor of 300.