Structural Damage Detection of Elevator Steel Plate Using GNARX Model

摘要

For elevator steel plate, the tiny crake is difficult to find out. However, the crake growth may cause fatality. Thus, it's crucial to detect structural damage of elevator steel plate. With the expression of GNARX model deduced, the modified Mahalanobis distance least square (MMDLS) is proposed for parameter estimation. Then, the structure pruning algorithm based on parameters' rate of standard deviation (SPRSD) is proposed for structure identification. With experimental data, GNARX model is applied to structural damage detection for elevator steel plate. The results show that the effect of structural damage detection of GNARX model is better than those of AR, ARX, GNAR models, which indicates the superiority of GNARX model applied to structural damage detection of elevator steel plate.