Search

Due to its importance in flow measurement and instrumentation, as well as its frequent application in differential pressure flowmeters, orifice discharge coefficient (C d) needs to be estimated precisely. In this study, different soft computing models (including multiple linear regression (MLR), group method of data handling (GMDH), multivariate adaptive regression splines (MARS), M5P tree model, and random forest (RF)) were employed for the first time in estimation of the C d value, and their respective prediction performances were analyzed statistically. Coefficient of correlation (CC), mean absolute error (MAE), root mean square error (RMSE), scattering index (SI), and Nash–Sutcliffe model efficiency coefficient (NSE) were used as the statistical indicators for validating the performance of each soft computing model. The statistical indicators approved the superiority of the RF model over the other models, while the MARS model also showed a competitive prediction potential over M5P, GMDH, and MLR models. The findings of this computational study clearly demonstrated that the implemented soft computing strategy had the capability to be used in precise estimation of the C d of the orifice meter, specifically, in situations where the measurement of the parameters in deterministic equation is not practically feasible. [Display omitted] • Differential pressure flowmeters were simulated using soft computing methodology. • MLR, GMDH, MARS, M5P, and RF were used to estimate C d for the first time. • Effectiveness of models was tested by CC, MAE, RMSE, SI, and NSE metrics. • RF outperformed other models, and MARS showed competitive prediction accuracy. • RF-based soft computation was of potential to be used in precise estimation of C d. [ABSTRACT FROM AUTHOR]

Owing to its importance in extraction of natural gas from underground gas storage as well as its crucial role in determination of final gas mixture in the production facilities of gas/oil industry, the dry content of wet gas mixture needs to be calculated precisely. The present study explores the potential of different soft-computing techniques in estimation of the dry gas flow rate (kg/h) (output variable) of wet gas mixture based on two input variables of wet gas flow rate (kg/h) and absolute gas humidity (g/m3). Decision tree-based methods (M5P tree, random forest (RF), random tree (RT), and reduced error pruning tree (REPT) models), kernel function-based approaches (Gaussian process regression (GPR) and support vector machines (SVM)), and non-parametric regression-based technique (multivariate adaptive regression splines (MARS)) were implemented for the first time to estimate the dry gas flow rate, and their respective prediction performances were analyzed statistically. Coefficient of correlation (CC), Nash–Sutcliffe efficiency (NSE), root mean square error (RMSE), mean absolute error (MAE), Legates and McCabe's index (LMI), and Willmott's Index (WI) were used as the statistical indicators for validating the performance of each soft-computing model. While M5P model (MAE = 122.2382 kg/h, RMSE = 580.5626 kg/h, CC = 0.9875 for the testing data set) was better than other tree-based models (MAE = 363.2802–542.6119 kg/h, RMSE = 871.9363–1025.3444 kg/h, CC = 0.9587–0.9706 for the testing data set) and MARS model (MAE = 128.0083 kg/h, RMSE = 622.9515 kg/h, CC = 0.9852 for the testing data set), the statistical indicators approved the superiority of the radial basis kernel function-based GPR model (GPR-RBKF) model (MAE = 163.3266 kg/h, RMSE = 483.1359 kg/h, CC = 0.9915 for the testing data set) over other implemented models in predicting the dry gas flow rate. The findings highlighted the potential of soft-computing methodologies in precise estimation of dry gas flow rate in wet gas mixture, particularly, in situations where the measurement of such parameters with traditional deterministic models is practically not possible. [Display omitted] • Soft-computing methods for estimation of dry content of wet gas mixture. • Performance evaluation of M5P, RF, RT, REPT, GPR, SVM, and MARS. • Superiority of GPR-RBKF over other models in terms of accuracy. • Superiority of M5P over other tree decision tree models in terms of error. • Higher precision of Gaussian processing with kernel-based regression vector. [ABSTRACT FROM AUTHOR]