Statistical Diagnosis for Damage Detection of Self-Learning Smart Structure.

Structural health monitoring is a noticeable technology for aged civil structures. The present paper proposes a new diagnostic tool for the structural health monitoring that employs a statistical diagnosis of self-learning method. Most of the structural health monitoring systems adopt parametric method based on modeling or non-parametric method such as artificial neural networks. The new statistic diagnosis method does not require the complicated modeling and a large number of data for the training of the artificial neural networks. In the present study, the proposed method is applied to detect pipe deflections due to plastic bending, which simulates disaster damage of gas pipes. Response surfaces among the measured natural frequencies of the pipes are produced at the initial stage and monitoring stages, and the difference of the response surfaces of the monitoring stages from the initial stage is statistically tested using F-test. As a result, the new method successfully diagnoses the damage without using modeling and a large number of data for training.