Computation of reservoir relative permeability curve based on multi-scale wavelet kernel extreme learning machine

In this paper, a novel calculating method on relative permeability curve is proposed based on multi-scale wavelet kernel extreme learning machine (MKELM). When kernel extreme learning machine (KELM) models for a complex system, the model might not realize high precision in multi-scale regression. The existing kernel ELM algorithms can only approximate model on the single-scale and may stick to local minima. To cope with those problems, a multi-scale wavelet ELM algorithm is proposed by using wavelet kernel. This algorithm is applied to calculate relative permeability curve. By comparing and analyzing the accuracy of relative permeability curve calculated by ELM, KELM and MKELM, the experimental results show that MKELM can improve the calculation accuracy obviously.