Eleni Chatzi | Swiss Federal Institute of Technology (ETH) (original) (raw)

Papers by Eleni Chatzi

Research paper thumbnail of Data-Driven Structural Health Monitoring and Diagnosis of Operating Wind Turbines

Research paper thumbnail of Data-Driven Tool for Structural Health Monitoring of Operating Wind Turbines

Scientific Journal of Civil Engineering

Research paper thumbnail of Parametric Model Order Reduction for Localized Nonlinear Feature Inclusion

NODYCON Conference Proceedings Series, 2021

Research paper thumbnail of Gaussian Process Time-Series Models for Structures under Operational Variability

Frontiers in Built Environment, Dec 8, 2017

Research paper thumbnail of Reduced order modeling of non-linear monopile dynamics via an AE-LSTM scheme

Frontiers in Energy Research

Non-linear analysis is of increasing importance in wind energy engineering as a result of their e... more Non-linear analysis is of increasing importance in wind energy engineering as a result of their exposure in extreme conditions and the ever-increasing size and slenderness of wind turbines. Whilst modern computing capabilities facilitate execution of complex analyses, certain applications which require multiple or real-time analyses remain a challenge, motivating adoption of accelerated computing schemes, such as reduced order modelling (ROM) methods. Soil structure interaction (SSI) simulations fall in this class of problems, with the non-linear restoring force significantly affecting the dynamic behaviour of the turbine. In this work, we propose a ROM approach to the SSI problem using a recently developed ROM methodology. We exploit a data-driven non-linear ROM methodology coupling an autoencoder with long short-term memory (LSTM) neural networks. The ROM is trained to emulate a steel monopile foundation constrained by non-linear soil and subject to forces and moments at the top o...

Research paper thumbnail of Nonlinear Reduced Order Modelling of Soil Structure Interaction Effects via LSTM and Autoencoder Neural Networks

Proceedings of the 13th International Workshop on Structural Health Monitoring

In the field of structural health monitoring (SHM), inverse problems which require repeated analy... more In the field of structural health monitoring (SHM), inverse problems which require repeated analyses are common. With the increase in the use of nonlinear models, the development of nonlinear reduced order modelling techniques is of paramount interest. Of considerable research interest, is the use of flexible and scalable machine learning methods which can learn to approximate the behaviour of nonlinear dynamic systems using input and output data. One such nonlinear system of interest, in the context of wind turbine structures, is the soil structure interaction (SSI) problem. Soil demonstrates strongly nonlinear behaviour with regards to its restoring force and has been shown to considerably influence the dynamic response of wind turbine structures. In this work, we demonstrate the application of a recently developed nonlinear reduced order modelling method, which leverages Autoencoder and LSTM neural networks, to a nonlinear soil structure interaction problem of a wind turbine mono...

Research paper thumbnail of On the Potential of Dynamic Sub-structuring Methods for Model Updating

Structural Health Monitoring 2019

Research paper thumbnail of An Integrated Monitoring Strategy for Current Condition Assessment of Historic Bridges

XI International Conference on Structural Dynamics, 2020

Research paper thumbnail of Vibration-based structural performance assessment via output only sub-Nyquist/ compressive wireless sensor data

This paper assesses two different approaches for efficient output - only Vibration - based Struct... more This paper assesses two different approaches for efficient output - only Vibration - based Structural Health Monitoring (V - SHM) in large - scale civil engineering structures, promoting the use of dense arrays of low - power wireless sensors. Firstly, a non - uniform deterministic sub - Nyquist multi - coset sampling scheme is considered to acquire ambient stationary structural response signals. This sampling scheme is coupled with a power spectrum blind sampling technique along with the frequency domain decomposition algorithm of operational modal analysis to obtain structural modal properties. This is accomplished without necessitating either signal reconstruction in the time - domain or signal sparsity assumption . Secondly, a spectro-temporal compressive sensing approach is considered applicable to cases where sign al reconstruction in time - domain is desired. The latter approach considers non-uniform in time random sampling at sub - Nyquist average rates informed by prior kno...

Research paper thumbnail of Cost-effective vibration based detection of wind turbine blade icing from sensors mounted on the tower

Ice throw from the blades of operational wind turbines is a safety concern when wind turbines are... more Ice throw from the blades of operational wind turbines is a safety concern when wind turbines are installed in a densely built environment. While several commercial solutions exist to detect icing or prevent ice build up altogether, there is still a desire for a more effective low-cost solution. In this contribution a previously instrumented onshore wind turbine is used to proof the concept of a novel ice detection strategy. In this new approach an accelerometer in the tower is used to infer the presence of icing on the rotor. Two approaches are investigated one based on natural frequencies extracted from operational modal analysis and the second based on Gaussian Process time series models.

Research paper thumbnail of On the Active Vibration Control of Nonlinear Uncertain Structures

Applied and Computational Mechanics, 2020

This study proposes an active nonlinear control strategy for effective vibration mitigation in no... more This study proposes an active nonlinear control strategy for effective vibration mitigation in nonlinear dynamical systems characterized by uncertainty. The proposed scheme relies on the coupling of a Bayesian nonlinear observer, namely the Unscented Kalman Filter (UKF) with a two-stage control process. The UKF is implemented for adaptive joint state and parameter estimation, with the estimated states and parameters passed onto the controller. The controller comprises a first task of feedback linearization, allowing for subsequent integration of any linear control strategy, such as addition of damping, LQR control, or other, which then operates on the linearized state equations. The proposed framework is validated on a Duffing oscillator characterized by light damping and an uncertain nonlinear parameter under harmonic and stochastic disturbance. The demonstrated performance suggests that the proposed Bayesian approach offers a competitive approach for active vibration suppression i...

Research paper thumbnail of Heterogeneous Sensor Fusion for Reducing Uncertainty in Structural Health Monitoring

Proceedings of the 1st International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECOMP 2015), 2015

Research paper thumbnail of Measuring sub-mm structural displacements using QDaedalus: a digital clip-on measuring system developed for total stations

Research paper thumbnail of Structural identification with physics-informed neural ordinary differential equations

Journal of Sound and Vibration, 2021

Research paper thumbnail of Fatigue Assessment in Steel Railway Bridges using Output Only Vibration Measurements and Partial Structural Information

Civil-comp proceedings, Mar 10, 2016

Research paper thumbnail of Optimal Sensor Placement for the Modal Identification of an Innovative Timber Structure

Research paper thumbnail of House of natural resources – modal vibration tests and long-term monitoring

This report presents an innovative case-study building implementing four novel structural systems... more This report presents an innovative case-study building implementing four novel structural systems implementing innovative hardwood materials. The main structural system is a post-tensioned timber frame with columns in ash glulam and beams in hybrid ash-spruce glulam. The beams and columns are connected via a horizontal post-tensioning cable, which creates a semi-rigid moment frame. Furthermore, three different innovative slab systems are implemented inside the building, first a timber-concrete composite slab with beech laminated veneer lumber (LVL), then a hollow-core timber-concrete composite slab and third a bi-axial timber slab with beech cross-laminated timber (CLT) and beech LVL lamellas. Test results from modal vibration tests during four different construction stages are presented. The results showcase the influence of the different structural systems on the dynamic properties of the building. Fundamental frequencies, damping ratios and mode shapes are derived from different sensor systems. Additionally, the monitoring data from a seven-year long monitoring campaign are presented. The most relevant physical quantities presented are for example, the long-term development of the post-tension force within the post-tensioned timber frame, or the long-term deformations of the bi-axial timber slab. The monitoring campaign illustrates that the building's long-term performance is generally equal to the long-term predictions, except for the deformations of the bi-axial beech slab, which will be investigated further

Research paper thumbnail of Exploring Topology Optimization on Hierarchical Meshes by Scaled Boundary Finite Element Method

Research paper thumbnail of Improving the Accuracy of Stress Intensity Factors Obtained by the Scaled Boundary Finite Element Method on Hierarchical Meshes

Research paper thumbnail of Input-state-parameter estimation of structural systems from limited output information

Mechanical Systems and Signal Processing, Jul 1, 2019

Research paper thumbnail of Data-Driven Structural Health Monitoring and Diagnosis of Operating Wind Turbines

Research paper thumbnail of Data-Driven Tool for Structural Health Monitoring of Operating Wind Turbines

Scientific Journal of Civil Engineering

Research paper thumbnail of Parametric Model Order Reduction for Localized Nonlinear Feature Inclusion

NODYCON Conference Proceedings Series, 2021

Research paper thumbnail of Gaussian Process Time-Series Models for Structures under Operational Variability

Frontiers in Built Environment, Dec 8, 2017

Research paper thumbnail of Reduced order modeling of non-linear monopile dynamics via an AE-LSTM scheme

Frontiers in Energy Research

Non-linear analysis is of increasing importance in wind energy engineering as a result of their e... more Non-linear analysis is of increasing importance in wind energy engineering as a result of their exposure in extreme conditions and the ever-increasing size and slenderness of wind turbines. Whilst modern computing capabilities facilitate execution of complex analyses, certain applications which require multiple or real-time analyses remain a challenge, motivating adoption of accelerated computing schemes, such as reduced order modelling (ROM) methods. Soil structure interaction (SSI) simulations fall in this class of problems, with the non-linear restoring force significantly affecting the dynamic behaviour of the turbine. In this work, we propose a ROM approach to the SSI problem using a recently developed ROM methodology. We exploit a data-driven non-linear ROM methodology coupling an autoencoder with long short-term memory (LSTM) neural networks. The ROM is trained to emulate a steel monopile foundation constrained by non-linear soil and subject to forces and moments at the top o...

Research paper thumbnail of Nonlinear Reduced Order Modelling of Soil Structure Interaction Effects via LSTM and Autoencoder Neural Networks

Proceedings of the 13th International Workshop on Structural Health Monitoring

In the field of structural health monitoring (SHM), inverse problems which require repeated analy... more In the field of structural health monitoring (SHM), inverse problems which require repeated analyses are common. With the increase in the use of nonlinear models, the development of nonlinear reduced order modelling techniques is of paramount interest. Of considerable research interest, is the use of flexible and scalable machine learning methods which can learn to approximate the behaviour of nonlinear dynamic systems using input and output data. One such nonlinear system of interest, in the context of wind turbine structures, is the soil structure interaction (SSI) problem. Soil demonstrates strongly nonlinear behaviour with regards to its restoring force and has been shown to considerably influence the dynamic response of wind turbine structures. In this work, we demonstrate the application of a recently developed nonlinear reduced order modelling method, which leverages Autoencoder and LSTM neural networks, to a nonlinear soil structure interaction problem of a wind turbine mono...

Research paper thumbnail of On the Potential of Dynamic Sub-structuring Methods for Model Updating

Structural Health Monitoring 2019

Research paper thumbnail of An Integrated Monitoring Strategy for Current Condition Assessment of Historic Bridges

XI International Conference on Structural Dynamics, 2020

Research paper thumbnail of Vibration-based structural performance assessment via output only sub-Nyquist/ compressive wireless sensor data

This paper assesses two different approaches for efficient output - only Vibration - based Struct... more This paper assesses two different approaches for efficient output - only Vibration - based Structural Health Monitoring (V - SHM) in large - scale civil engineering structures, promoting the use of dense arrays of low - power wireless sensors. Firstly, a non - uniform deterministic sub - Nyquist multi - coset sampling scheme is considered to acquire ambient stationary structural response signals. This sampling scheme is coupled with a power spectrum blind sampling technique along with the frequency domain decomposition algorithm of operational modal analysis to obtain structural modal properties. This is accomplished without necessitating either signal reconstruction in the time - domain or signal sparsity assumption . Secondly, a spectro-temporal compressive sensing approach is considered applicable to cases where sign al reconstruction in time - domain is desired. The latter approach considers non-uniform in time random sampling at sub - Nyquist average rates informed by prior kno...

Research paper thumbnail of Cost-effective vibration based detection of wind turbine blade icing from sensors mounted on the tower

Ice throw from the blades of operational wind turbines is a safety concern when wind turbines are... more Ice throw from the blades of operational wind turbines is a safety concern when wind turbines are installed in a densely built environment. While several commercial solutions exist to detect icing or prevent ice build up altogether, there is still a desire for a more effective low-cost solution. In this contribution a previously instrumented onshore wind turbine is used to proof the concept of a novel ice detection strategy. In this new approach an accelerometer in the tower is used to infer the presence of icing on the rotor. Two approaches are investigated one based on natural frequencies extracted from operational modal analysis and the second based on Gaussian Process time series models.

Research paper thumbnail of On the Active Vibration Control of Nonlinear Uncertain Structures

Applied and Computational Mechanics, 2020

This study proposes an active nonlinear control strategy for effective vibration mitigation in no... more This study proposes an active nonlinear control strategy for effective vibration mitigation in nonlinear dynamical systems characterized by uncertainty. The proposed scheme relies on the coupling of a Bayesian nonlinear observer, namely the Unscented Kalman Filter (UKF) with a two-stage control process. The UKF is implemented for adaptive joint state and parameter estimation, with the estimated states and parameters passed onto the controller. The controller comprises a first task of feedback linearization, allowing for subsequent integration of any linear control strategy, such as addition of damping, LQR control, or other, which then operates on the linearized state equations. The proposed framework is validated on a Duffing oscillator characterized by light damping and an uncertain nonlinear parameter under harmonic and stochastic disturbance. The demonstrated performance suggests that the proposed Bayesian approach offers a competitive approach for active vibration suppression i...

Research paper thumbnail of Heterogeneous Sensor Fusion for Reducing Uncertainty in Structural Health Monitoring

Proceedings of the 1st International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECOMP 2015), 2015

Research paper thumbnail of Measuring sub-mm structural displacements using QDaedalus: a digital clip-on measuring system developed for total stations

Research paper thumbnail of Structural identification with physics-informed neural ordinary differential equations

Journal of Sound and Vibration, 2021

Research paper thumbnail of Fatigue Assessment in Steel Railway Bridges using Output Only Vibration Measurements and Partial Structural Information

Civil-comp proceedings, Mar 10, 2016

Research paper thumbnail of Optimal Sensor Placement for the Modal Identification of an Innovative Timber Structure

Research paper thumbnail of House of natural resources – modal vibration tests and long-term monitoring

This report presents an innovative case-study building implementing four novel structural systems... more This report presents an innovative case-study building implementing four novel structural systems implementing innovative hardwood materials. The main structural system is a post-tensioned timber frame with columns in ash glulam and beams in hybrid ash-spruce glulam. The beams and columns are connected via a horizontal post-tensioning cable, which creates a semi-rigid moment frame. Furthermore, three different innovative slab systems are implemented inside the building, first a timber-concrete composite slab with beech laminated veneer lumber (LVL), then a hollow-core timber-concrete composite slab and third a bi-axial timber slab with beech cross-laminated timber (CLT) and beech LVL lamellas. Test results from modal vibration tests during four different construction stages are presented. The results showcase the influence of the different structural systems on the dynamic properties of the building. Fundamental frequencies, damping ratios and mode shapes are derived from different sensor systems. Additionally, the monitoring data from a seven-year long monitoring campaign are presented. The most relevant physical quantities presented are for example, the long-term development of the post-tension force within the post-tensioned timber frame, or the long-term deformations of the bi-axial timber slab. The monitoring campaign illustrates that the building's long-term performance is generally equal to the long-term predictions, except for the deformations of the bi-axial beech slab, which will be investigated further

Research paper thumbnail of Exploring Topology Optimization on Hierarchical Meshes by Scaled Boundary Finite Element Method

Research paper thumbnail of Improving the Accuracy of Stress Intensity Factors Obtained by the Scaled Boundary Finite Element Method on Hierarchical Meshes

Research paper thumbnail of Input-state-parameter estimation of structural systems from limited output information

Mechanical Systems and Signal Processing, Jul 1, 2019