PSI - Issue 17

Ahmed Belmokre et al. / Procedia Structural Integrity 17 (2019) 698–703 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

701

4

Fig. 2. Evolution of water temperature at elevation 175 m above mean sea level.

3.2. Results

The predicted and recorded low rates at flowmeters FL-175 and FR-175 are depicted in Fig. 3. The performance of both models is evaluated by the root mean square error (RMSE) and the mean absolute error (MAE):

1 N N = i



(

) 2

RMSE

X Y −

=

(4)

i

i

1

1 N N = i



MAE

X Y −

=

(5)

i

i

1

Where X i and Y i are the measured and predicted value respectively, and N . is the number of observations. The performance of both models is listed in Table 1. The RFR model provides better estimates than the SVR model. The better performance of the RFR model could be explained thanks its ability to handle nonlinear relationships between variables, since the seepage flow rate in dams is a nonlinear process.

Table 1. Performance of RF and SVR models Drain Model RMSE ( l/s )

MAE ( l/s )

RFR SVR RFR SVR

0.0629 0.0969 0.1332 0.1532

0.0327 0.0706 0.0901 0.0971

FL-175

FR-175

(a)