|
| Application of Neural Network Techniquein Individual Strain Life Monitoring |
| Received:August 11, 2018 Revised:December 25, 2018 |
| View Full Text View/Add Comment Download reader |
| DOI：10.7643/ issn.1672-9242.2018.12.014 |
| KeyWord:strain monitoring neural networks load model damage |
| Author | Institution |
| GU Yu-xuan |
Shenyang Aircraft Design and Research Institute, Shenyang , China |
| SUI Fu-cheng |
Shenyang Aircraft Design and Research Institute, Shenyang , China |
| SONG En-peng |
Shenyang Aircraft Design and Research Institute, Shenyang , China |
|
| Hits: |
| Download times: |
| Abstract: |
| Objective To solve the error caused by nonlinear problems in the single strain monitoring load model. Methods Based on the technique of artificial neural network and in combination with the finite element model of one airplane’s typical box-section, a neural network load model used for ISM was established to validate its accuracy with random samples. Results The predictive value of BP neural network was close to the measured values of load model; and its prediction result was better than that of the multiple linear regression model. Conclusion The BP neural network load model is suitable for ISM. In addition, because of its high prediction accuracy, the damage of key parts could be monitored more accurately with the usage of neural network load model. |
| Close |