An improved fuzzy neural network for permeability estimation from wireline logs in a petroleum reservoir

Reservoir permeability estimation from wireline logs is the most difficult task for petrophysicists. Many studies have shown that the backpropagation neural network (BPNN) is the most promising tool to date, because of its ability to learn and generalise. This paper presents an improved fuzzy neural network (FNN) to solve the same problem. In the example presented, this model is stable with fast convergence and gives smaller error compared to BPNN and previous FNN methods.