Research Article
Applying a Cerebellar Model Articulation Controller Neural Network to a Photovoltaic Power Generation System Fault Diagnosis
Table 8
Test data with an additional error and under 300–1000 W/m2 irradiation and PV module surface temperatures between 31 and 40°C.
| Test no. | (W) | (V) | (A) | (V) | Temperature (°C) | Irradiation (W/m2) | Known fault type |
| 1 | 825.96
*
| 341.92
*
| 2.8 | 372.56 | 31.5 | 300 | PF1 | 2 | 1195.23 | 313.34 | 3.04
*
| 415.73
*
| 32.5 | 400 | PF1 | 3 | 1399.53 | 289.25 | 3.87
*
| 420
*
| 33.5 | 500 | PF2 | 4 | 1695.32 | 319.68
*
| 5.83 | 423.42
*
| 34.5 | 600 | PF2 | 5 | 1670.43
*
| 255.11 | 5.51
*
| 387.41 | 35.5 | 700 | PF3 | 6 | 2013.52 | 282.55
*
| 6.27
*
| 389.52 | 36.5 | 800 | PF3 | 7 | 1949.64 | 219.32 | 7.11
*
| 430.38
*
| 37.5 | 900 | PF4 | 8 | 2055.82
*
| 217.65 | 7.95
*
| 392.78 | 38.5 | 1000 | PF4 | 9 | 761.64 | 295.88
*
| 2.83 | 359.45
*
| 39.5 | 300 | PF5 | 10 | 1011.02
*
| 305.59
*
| 3.83 | 336.43 | 31.5 | 400 | PF5 | 11 | 1124.59
*
| 244.95 | 4.83 | 327.19
*
| 32.5 | 500 | PF6 | 12 | 1362.03
*
| 270.81
*
| 5.82 | 299.85 | 33.5 | 600 | PF6 | 13 | 1441.41 | 209.59 | 5.5
*
| 284.55
*
| 34.5 | 700 | PF7 | 14 | 1567.73
*
| 228.6
*
| 7.94 | 260.08 | 35.5 | 800 | PF7 | 15 | 2251.15 | 276.31
*
| 7.16
*
| 348.29 | 36.5 | 900 | PF8 | 16 | 2373.74
*
| 253.42 | 9.86 | 384.59
*
| 37.5 | 1000 | PF8 | 17 | 602.75 | 232.45
*
| 2.28
*
| 327.28 | 38.5 | 300 | PF9 | 18 | 785.31
*
| 234.09
*
| 3.88 | 332.09 | 39.5 | 400 | PF9 | 19 | 1057.41 | 238.64
*
| 4.87 | 327.62
*
| 31.5 | 500 | PF10 | 20 | 1281.44 | 218.22 | 4.69
*
| 330.25
*
| 32.5 | 600 | PF10 |
| Variance | −5% | +10% | −20% | +10% |
|
|
*Indicates the addition of variance in the test data.
|