28.02.2014 Views

The Development of Neural Network Based System Identification ...

The Development of Neural Network Based System Identification ...

The Development of Neural Network Based System Identification ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

5.4 OFF-LINE BASED SYSTEM IDENTIFICATION FOR ELMAN NETWORK 143<br />

to the range <strong>of</strong> measurement between −1 rad/s to 1 rad/s. This indicates the imprecise<br />

and high level <strong>of</strong> uncertainty to produce prediction that represent the real output values.<br />

Table 5.9 <strong>The</strong> average RMSE for various noise levels applied to optimum weights <strong>of</strong> the modified<br />

Elman network (4 hidden neurons).<br />

Standard Deviation<br />

<strong>of</strong> Noise<br />

0.01<br />

0.05<br />

0.1<br />

0.3<br />

0.5<br />

0.7<br />

Test Error Statistics<br />

<strong>System</strong><br />

Responses<br />

RMSE RMSE (%) R2 NMCIW (%)<br />

p 0.0658 15.6337 0.9753 0.0876<br />

q 0.0174 5.4138 0.9966 0.2425<br />

p 0.0666 15.8336 0.9747 0.5408<br />

q 0.0163 5.0892 0.9970 1.1632<br />

p 0.0661 15.7048 0.9751 0.8776<br />

q 0.0188 5.8519 0.9961 2.5763<br />

p 0.0703 16.7037 0.9718 2.7539<br />

q 0.0699 21.8101 0.9453 6.8574<br />

p 0.0713 16.9354 0.9710 5.9380<br />

q 0.0745 23.2384 0.9379 13.1676<br />

p 0.0985 23.4007 0.9447 12.6861<br />

q 0.0459 14.3050 0.9765 22.4513

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!