Research Article
Damage Localization and Quantification of Truss Structure Based on Electromechanical Impedance Technique and Neural Network
Table 1
Expected outputs and actual outputs of testing samples.
| Rod 1 | Damage states | 2 | 4 | 8 | Output | Expected | 0 | 0 | 1–3 | 0 | 0 | 3–5 | 0 | 0 | 7–9 | Actual | 0.0004 | 0.0001 | 1.6398 | 0.0004 | 0.0007 | 3.8930 | 0.0023 | 0.0659 | 7.3636 |
| Rod 2 | Damage states | 1 | 3 | 7 | Output | Expected | 0 | 1 | 0–2 | 0 | 1 | 2–4 | 0 | 1 | 6–8 | Actual | 0.0062 | 0.9990 | 1.0841 | 0.0061 | 0.9858 | 2.3948 | 0.0031 | 0.9771 | 6.5686 |
| Rod 3 | Damage states | 2 | 7 | 9 | Output | Expected | 1 | 0 | 1–3 | 1 | 0 | 6–8 | 1 | 0 | 8–10 | Actual | 0.9977 | 0.0002 | 1.9314 | 0.9959 | 0.0085 | 8.0894 | 0.9956 | 0.0456 | 8.8745 |
| Rod 4 | Damage states | 1 | 5 | 8 | Output | Expected | 1 | 1 | 0–2 | 1 | 1 | 4–6 | 1 | 1 | 7–9 | Actual | 0.9946 | 0.9975 | 1.6656 | 0.9960 | 0.9982 | 5.2482 | 0.9960 | 0.9173 | 8.5058 |
|
|