Modeling the Natural Gas Compressibility Factor through Adaptive Neuro-Fuzzy Inference System

Accurate determination of the natural gas compressibility factor is crucial for reservoir simulation and material balance computations in petroleum engineering. the datadriven ai techniques, like artificial neural networks, fuzzy systems, and neuro-fuzzy systems, are gaining momentum in estimating fluid properties. an adaptive neuro-fuzzy inference system (anfis) is applied here to develop a model to estimate the compressibility factor of two natural gas types. the takagi-sugeno fuzzy inference system serves as the foundation for constructing the anfis model, where the triangular membership functions are applied. the training data consists of 80% of the available data selected randomly, and the remaining 20% is applied in testing. this developed model is of high accuracy in estimating the compressibility factors of natural gas types, with an average absolute relative deviation of 0.05% and a maximum absolute relative deviation of 0.55% difference between the estimated and experimental value data. comparing the findings here with the correlations indicates that the anfis model in terms of accuracy outperforms its counterparts in this realm.