Symbols: black square, relative activity ; gray square, conversion after 24-h reaction.
Predicted parameters yield an average RMSD (root mean square relative deviation) of 40% from vapor pressure data.
Root mean square relative error (RMSRE) was calculated to indicate the prediction performance of the obtained models.
For the MLR model, the prediction root mean square relative error (RMSRE) of leave one out cross validation and external validation is 2.82 and 2.95, respectively.
They also include modes of predication, related to Aristotle's ten categories, such as substantial (e.g., horse), qualitative (e.g., brown), quantitative (e.g., square), relative and so on.
The comparison results show that the sum of square relative errors of the model obtained by the DNA-GA is the least of the test models, and its prediction is in good agreement with the practical operating data.
These five error parameters are given in Table 2 (the absolute error, the minimum and maximum absolute error, the correlation coefficient, standard deviation and finally the root mean square relative error).
Four performance criteria are used in this study to assess the goodness of fit of the models, which are: root mean square error (RMSE), mean absolute error (MAE), mean square relative error (MSRE), and coefficient of determination (R 2) (further discussed by Ghorbani et al. 2012).
Zhao and Wang (2007) suggested using "optimal smooth coefficient", try any combination of three smoothing coefficient using exhaustive algorithm in EViews 6.0 software, the corresponding smoothing coefficient is selected as "optimal smoothing coefficient" when the sum of square relative error reaches the minimum, which is expressed as Eq. (5).
In the derivation of the equation of motion, non-linearities due to large lateral displacements as well as rotation, drag wind and wave forces, which are assumed to be proportional to the square relative velocity between the wind/wave and the mast, are considered.
