Summary
The finite-difference methods have been widely utilized in seismic wave numerical modelling, seismic imaging and inversion.we can truncate spatial convolutional counterpart of the pseudospectral methods to get finite-difference operators using truncated windows, and the properties of truncated windows in amplitude responses play a major role in finite-difference approximation, narrower main lobe and larger attenuation of side lobe bring higher accuracy of finite-difference approximation, we propose an optimized finite-difference operators based on Chebyshev auto-convolution combined window which has narrow main lobe and large attenuation of side lobe in its amplitude response. Through adjusting three parameters of auto-convolution combined window, visually and intuitively control the properties of main lobe and side lobe to adjust accuracy of finite-difference approximation. Numerical dispersion analysis and elastic wave numerical modelling demonstrate that our finite-difference operators have higher accuracy than the conventional finite-difference operators and can more efficiently suppress numerical dispersion under the same condition.
