Time-Domain Structural Damage Identification Using Ensemble Bagged Trees and Evolutionary Optimization Algorithms

Abstract

This paper presents a two-step vibration-based strategy for damage identification of framed structures using ensemble bagged trees known as a well-known supervised machine learning (ML) paradigm in conjunction with evolutionary optimization algorithms. The proposed model incorporates the actual response, wavelet coefficients, and wavelet energy to extract damage-sensitive features from the time-domain of the measured and simulated signals. Unlike available studies in this scope, the key objective of this research is to identify damage with a localization precision down to a single structural member, rather than limiting the evaluation to the group of elements. In order to increase the training performance in contributing to extremely large datasets with numerous class labels, the proposed strategy involves the artificial generation of features. Additionally, a modified genetic algorithm is proposed for fast damage localization. It is deduced that the damage locations are confidently detected within a fast computational time. Subsequently, damage identification is followed by the application of evolutionary optimization algorithms. For comparison purpose, the employment of the water cycle optimization algorithm (WCA) is comparatively investigated with the other three state-of-the-art optimizers, i.e., particle swarm optimization (PSO), imperialist competitive algorithm (ICA), and differential evolution algorithm (DE). The numerical and experimental validation studies evidence satisfactorily reliable identification results with no false detection in dealing with multiple damage scenarios in large-scale and real-world applications. It is concluded that developing the most damage-sensitive features and using the proposed data fusion strategy lead to informative features with a reasonably small size and significantly improve the ML performance.

On the influence of rotor nacelle assembly modelling on the computed eigenfrequencies of offshore wind turbines

“…This approach is done due to lack of available information on the blade properties from turbine manufacturers and to manage the model complexity, but fails to take the flexibility of the blades into account. The impact of including flexible blade models has been subject of previous studies in [8], [12], and [13]. While in [8] no difference between flexible and rigid blade models were noticed, other studies focusing on the influence of rotor nacelle assembly modelling in [12] and [13] did find a significant influence of the blade modelling on the second and higher tower modes, see Figure 2.…”

“…The impact of including flexible blade models has been subject of previous studies in [8], [12], and [13]. While in [8] no difference between flexible and rigid blade models were noticed, other studies focusing on the influence of rotor nacelle assembly modelling in [12] and [13] did find a significant influence of the blade modelling on the second and higher tower modes, see Figure 2. 1 st mode Ali et.…”

“…The existence of a mismatch between computed (or designed) and as-built eigenfrequencies for offshore wind turbines with monopile foundations is well known [1]. Improvements of the soil-structure interaction modelling through the PISA project [2] have reduced this mismatch, especially for the first tower mode [5], [6], [7], [8]. However, the mismatch between measurements and computations still exists for the second tower mode; both for jacket [9], [10] and monopile foundations [6], [11].…”

“…With the exception of [5] and [8], a common feature of the structural models used in the aforementioned studies is the use of a lumped mass approach to model the rotor nacelle assembly. This approach is done due to lack of available information on the blade properties from turbine manufacturers and to manage the model complexity, but fails to take the flexibility of the blades into account.…”

“…To mimic the free to rotate rotor in idling condition, the torsional stiffness of the rigid beam element between the rotor and master node is set to a value near zero. A similar approach has been previously used in [8].…”