The absolute fraction of variance (R2) is 0.9999 which can be considered as very promising.
In addition, the absolute fraction of variance (R2) and mean absolute percentage error (MAPE) are calculated using Eqs.
Accuracy of ANN model in terms of the root absolute fraction of variance (R) and the mean squared error (MSE) are evaluated.
Scatter diagrams and two statistical criteria, i. e. absolute fraction of variance (R2) and mean relative error (MRE), were used to evaluate the prediction performance of the developed model.
Using the local DCAs as representative variables allows the retention of a larger fraction of variance from each site, removes any subjectivity from variable selection and retains the potential for observing multiple, coherent signals from within and between each dataset.
Scatter plots and statistical criteria of "absolute fraction of variance (R2)", and "mean relative error (MRE)" were used to evaluate the prediction performance and universality of the developed models.
For the torque testing data, root mean squared-error (RMSE), fraction of variance (R2) and mean absolute percentage error (MAPE) were found to be 0.9017%, 0.9920% and 7.2613%, respectively.
Absolute fraction of variance and root mean square error of 0.99 and 0.008, in training and testing phases of the model were achieved showing the relatively high accuracy of the proposed ANFIS model.
For this number level, after the training, it is found that root-mean squared (RMS) value is 0.0047, and absolute fraction of variance (R2) value is 0.9999 and coefficient of variation in percent (cov) value is 0.1363.
The comparison of the absolute fraction of variance (R2) (0.998 and 0.961), the root mean square error (RMSE) (0.105 kg/h and 0.489 kg/h) and the mean absolute percentage error (MAPE) (0.954% and 4.75%) demonstrated the result for combination of dimensional analysis and Adaptive Neuro-Fuzzy Inference System and dimensionless correlation model predictions respectively.
The output of the ANN models were found to be in agreement with experimental data with an absolute fraction of variance (R2) higher than 0.99 in the cases of CH4 and CO models and higher than 0.98 in the case of CO2 and H2 model.
