A novel efficient model for gas compressibility factor based on GMDH network

With the development of the low-power flowmeters, it is urgent and crucial to calculate the gas compression factor (Z-factor) in real time. However, the traditional estimation methods for Z-factor such as AGA8-92DC, needed a long calculation time and were difficult to be applied to the low-power embedded-based gas flowmeters. The other empirical correlations also had large errors. To solve this problem, we proposed a novel model for quick calculation of gas Z-factor based on Group Method of Data Handling(GMDH) network. The accuracy and reliability of this model were verified under different Tpr and Ppr. Compared with other empirical correlations, our model has the lowest root mean sum of squares of the errors (RMSE) of 0.0066 and mean absolute percentage error (MAPE) of 0.5615%, which is only 1/13–2/5 of MAPE calculated by the other correlations. The results show that our model has higher accuracy. Moreover, our model avoids lots of complex exponential and logarithmic operations, so it is especially useful for real-time Z-factor calculation of the low-power flowmeters