Multi-Constrained Seismic Multi-Parameter Full Waveform Inversion Based on Projected Quasi-Newton Algorithm

The multi-parameter full waveform inversion (FWI) that integrates velocity and density can make use of the kinematic dynamic information measured data to reconstruct underground model. However, it faces problems crosstalk between multiple parameters strong nonlinearity. This research proposes a multi-constrained, FWI framework based on projected quasi-Newton algorithm. introduce types prior geological information, which effectively improve problem inversion. Additionally, method eliminate phenomenon further convergence speed. Taking 1994BP model as an example, results show has faster speed than spectral gradient method, reduces parameters; constraint sets are uniquely onto intersection ensure estimated values meet constraints. We also experiment with overthrust model, shows we proposed accuracy good adaptability. be compatible more obtain conforms understanding great potential in seismic exploration.