Table 4. Comparison of Different Structures in ANN Models.
| no. | algorithms | hidden nodes | max. error % | training MSE × 102 | testing MSE × 102 |
|---|---|---|---|---|---|
| 1 | BR | 10 | 46% | 0.162 | 5.245 |
| 2 | BR | 20 | 40% | 0.170 | 3.159 |
| 3 | BR | 20, 10a | 44% | 0.075 | 1.221 |
| 4 | BR | 20, 20a | 25% | 0.074 | 1.200 |
| 5 | BR | 20, 30a | 22% | 0.058 | 0.866 |
| 6 | BR | 20, 40a | 22% | 0.071 | 0.950 |
| 7 | BR | 20, 50a | 31% | 0.063 | 2.070 |
It is a double-hidden layer network.