Damage Identification of a Steel Frame Based on Integration of Time Series and Neural Network under Varying Temperatures

Abstract

The effect of varying temperatures is one of the most important challenges of vibration-based damage identification due to its bigger effects on the structural response than the damage itself. This study presents a methodology incorporating the autoregressive (AR) time series model with two-step artificial neural networks (ANNs) to identify damage under temperature variations. AR coefficients, which are extracted by fitting the AR models to acceleration responses, are however sensitive to temperature changes, resulting in false diagnoses. Thus, two-step ANN models with the inputs of difference in AR coefficients are utilized to compensate the detrimental temperature variations. Finite element (FE) models of a steel-braced frame structure, simulating several damage scenarios with different damage locations and severities at fluctuating temperatures, are used to verify the effectiveness and reliability of this approach. Numerical results indicate that the proposed approach could successfully recognize, locate, and quantify damage by using output-only vibration and temperature data regardless of varying temperatures and noise perturbations.

Damage Localization in Composite Plates Using Wavelet Transform and 2-D Convolutional Neural Networks

“…The evaluation of structural systems with ML applications, such as wind turbines [ 11 ], automobiles [ 12 ], rotating machinery [ 13 ], and civil structures [ 14 , 15 , 16 , 17 ], has demonstrated their ability to perform reliable predictions. Recent advances in aerospace structure SHM using ML methods have demonstrated applications for impact location (e.g., using a 2-D CNN to detect the impact zone [ 18 ], a time-series signal rearranged as 2-D images for classifying both the structure’s status and the impact zone via CNN [ 19 ], and the impact coordinates using an Artificial Neural Network (ANN) [ 20 ]), damage localization in plate-like structures (using acceleration data [ 21 ] and simulating damage located in different locations with rearranged time-series signals to 2-D images [ 22 ], or using spectrograms by time-varying 1-D CNN [ 23 ]) and fatigue prognosis [ 24 ].…”

Damage Detection in Different Types of 3D Asymmetric Buildings Using Vibration Characteristics

“…Generally, the performance of a damage indicator or a damage identification technique is decided by structural type [6]. Structures that received the greatest research interest include beams [7,8], plate elements [9][10][11], trusses [12][13][14], steel frames [15,16], offshore platforms [17][18][19], and bridges [20][21][22].…”

Practical Implementation of Structural Health Monitoring in Multi-Story Buildings

“…Nonparametric and parametric methods yielded effective outputs for the time series method to detect global or local damage and precisely predict actual damage [151,152]. Huang et al [153] adopted an Autoregressive and an ANN model to predict variations in temperature to carry out a vibration-based damage identification process. FE models were used to demonstrate the reliability and effectiveness of this method.…”