Substructure Vibration NARX Neural Network Approach for Statistical Damage Inference

A damage detection approach is developed using nonlinear autoregressive with exogenous inputs (NARX) neural networks and a statistical inference technique. Within a large spatially extended dynamic system, an instrumented local substructure may be represented by a neural network, to predict the dynamic response of a given sensor from that of its neighbors. Without change in the system properties, the network prediction error will follow a stable statistical distribution. To infer damage, change in the prediction error variance as evaluated by the statistical inference standard F test is utilized as a sensitive indicator. Validation of the described procedure is undertaken using two experimental data sets (from the Los Alamos National Laboratory in Los Alamos, NM). Reduced stiffness and nonlinear response of a mass-spring system is documented in the first set, while joint damage in a frame structure is explored in the second. Favorable results are obtained in both cases with linear/nonlinear and single/multidamage patterns. Overall, the proposed framework may be particularly efficient for large spatially extended sensor network situations, where local condition assessment may be conducted based on the response of a few neighboring sensors.