Elsa Henriques
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Elsa Henriques has a total of 122 co-authors. Here are the top co-authors by number of shared publications:
- Paulo Peças
IDMEC, Instituto Superior Técnico, Universidade de Lisboa, 1049-001 Lisboa, Portugal
35 shared publications - Inês Ribeiro
IDMEC, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugal
21 shared publications - Arlindo Silva
Singapore University of Technology and Design
18 shared publications - Cairo Lucio Nascimento
Division of Electronic Engineering, Instituto Tecnológico de Aeronáutica, São José dos Campos, Brazil
7 shared publications - Luis Reis
University of Lisbon Lisbon Portugal
7 shared publications - Mihail Fontul
IBER-OLEFF Componentes Técnicos em Plástico Sa Parque Industrial Manuel da Mota, Lt 10/18, 3100-354 Pombal, Portugal
7 shared publications
Latest Publications
- Open Access
- 5 November 2024
- Published by PHM Society in Annual Conference of the PHM Society
- Vol. 16 (1)
Using Explainable Artificial Intelligence to Interpret Remaining Useful Life Estimation with Gated Recurrent Unit
In engineering, prognostics can be defined as the estimation of the remaining useful life of a system given current and past health conditions. This field has drawn attention from research, industry, and government as this kind of technology can help improve efficiency and lower the costs of maintenance in a variety of technical applications. An approach to prognostics that has gained increasing attention is the use of data-driven methods. These methods typically use pattern recognition and machine learning to estimate the residual life of equipment based on historical data. Despite their promising results, a major disadvantage is that it is difficult to interpret this kind of methodologies, that is, to understand why a certain prediction of remaining useful life was made at a certain point in time. Nevertheless, the interpretability of these models could facilitate the use of data-driven prognostics in different domains such as aeronautics, manufacturing, and energy, areas where certification is critical. To help address this issue, we use Local Interpretable Model-agnostic Explanations (LIME) from the field of eXplainable Artificial Intelligence (XAI) to analyze the prognostics of a Gated Recurrent Unit (GRU) on the C-MAPSS data. We select the GRU as this is a deep learning model that a) has an explicit temporal dimension and b) has shown promising results in the field of prognostics and c) is of simplified nature compared to other recurrent networks. Our results suggest that it is possible to infer the feature importance for the GRU both globally (for the entire model) and locally (for a given RUL prediction) with LIME.
- 5 November 2024
- Published by PHM Society in Annual Conference of the PHM Society
- Vol. 16 (1)
Open AccessUsing Explainable Artificial Intelligence to Interpret Remaining Useful Life Estimation with Gated Recurrent Unit
In engineering, prognostics can be defined as the estimation of the remaining useful life of a system given current and past health conditions. This field has drawn attention from research, industry, and government as this kind of technology can help improve efficiency and lower the costs of maintenance in a variety of technical applications. An approach to prognostics that has gained increasing attention is the use of data-driven methods. These methods typically use pattern recognition and machine learning to estimate the residual life of equipment based on historical data. Despite their promising results, a major disadvantage is that it is difficult to interpret this kind of methodologies, that is, to understand why a certain prediction of remaining useful life was made at a certain point in time. Nevertheless, the interpretability of these models could facilitate the use of data-driven prognostics in different domains such as aeronautics, manufacturing, and energy, areas where certification is critical. To help address this issue, we use Local Interpretable Model-agnostic Explanations (LIME) from the field of eXplainable Artificial Intelligence (XAI) to analyze the prognostics of a Gated Recurrent Unit (GRU) on the C-MAPSS data. We select the GRU as this is a deep learning model that a) has an explicit temporal dimension and b) has shown promising results in the field of prognostics and c) is of simplified nature compared to other recurrent networks. Our results suggest that it is possible to infer the feature importance for the GRU both globally (for the entire model) and locally (for a given RUL prediction) with LIME.
- Open Access
Economic Viability of Implementing Structural Health Monitoring Systems on the Support Structures of Bottom-Fixed Offshore Wind
Offshore wind (OSW) energy is a renewable source with strong prospects of development that may decisively contribute towards energy independence. Offshore wind is, however, not yet ubiquitously cost competitive, and frequently requires support schemes to finance its extensive capital requirements. Therefore, cost reduction strategies are necessary for the future development of offshore wind technologies. Even if structural health monitoring (SHM) systems are currently applied for the inspection of critical mechanical structures, they have not been the focus of research from offshore wind stakeholders. The main goal of this study is to evaluate the viability of SHM systems on the support structures of bottom-fixed offshore wind (BFOSW), alongside the impact of implementing these systems on life-cycle. Economic models are used to estimate the impact of implementing these systems, explained using a case-study of the Kaskasi farm in the German North Sea. General results indicate that installing SHM systems on the support structures of offshore wind can shift the maintenance strategies from preventive to predictive, allowing the intervals between inspections to be increased without a reduction on equipment availability. The greatest benefit is related with the possibility of extending the operational life of the farm.
Open AccessEconomic Viability of Implementing Structural Health Monitoring Systems on the Support Structures of Bottom-Fixed Offshore Wind
Offshore wind (OSW) energy is a renewable source with strong prospects of development that may decisively contribute towards energy independence. Offshore wind is, however, not yet ubiquitously cost competitive, and frequently requires support schemes to finance its extensive capital requirements. Therefore, cost reduction strategies are necessary for the future development of offshore wind technologies. Even if structural health monitoring (SHM) systems are currently applied for the inspection of critical mechanical structures, they have not been the focus of research from offshore wind stakeholders. The main goal of this study is to evaluate the viability of SHM systems on the support structures of bottom-fixed offshore wind (BFOSW), alongside the impact of implementing these systems on life-cycle. Economic models are used to estimate the impact of implementing these systems, explained using a case-study of the Kaskasi farm in the German North Sea. General results indicate that installing SHM systems on the support structures of offshore wind can shift the maintenance strategies from preventive to predictive, allowing the intervals between inspections to be increased without a reduction on equipment availability. The greatest benefit is related with the possibility of extending the operational life of the farm.
- Open Access
- 24 April 2020
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 8 (4) , 298
Soil Interaction and Grout Behavior for the NREL Reference Monopile Offshore Wind Turbine
Monopiles for offshore wind are the most used foundations by farm operators due to their low production costs, when compared to other bottom-fixed or floating foundations. In this research, a monopile foundation for offshore wind power was evaluated for its soil interaction and grout behavior, and an appropriate numerical model for the structural analysis of the foundation and tower was developed. FAST 8, an aero-hydro-servo-elastic numerical code developed by NREL, was used to obtain the loads applied on the supporting structures. These loads were pre-processed before they were inputted on the finite element model, developed using the finite element software ANSYS. The considered conical grout connection, which connects the monopile to the transition piece through friction, was modeled under a changing-status nonlinearity condition. To model the soil–pile interaction, a p-y model was applied using the ANSYS APDL commands. Static, modal, and transient structural analyses were produced to study the structure suitability for its use on offshore environments. Different soil interactions were modeled, and their results were then compared within the transient and modal analysis, indicating that the angle of the grout connection strongly affects the loading conditions on the grout. Moreover, scouring affects the dynamic behavior of the overall supporting structures, thus protection against this phenomenon is suggested.
- 24 April 2020
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 8 (4) , 298
Open AccessSoil Interaction and Grout Behavior for the NREL Reference Monopile Offshore Wind Turbine
Monopiles for offshore wind are the most used foundations by farm operators due to their low production costs, when compared to other bottom-fixed or floating foundations. In this research, a monopile foundation for offshore wind power was evaluated for its soil interaction and grout behavior, and an appropriate numerical model for the structural analysis of the foundation and tower was developed. FAST 8, an aero-hydro-servo-elastic numerical code developed by NREL, was used to obtain the loads applied on the supporting structures. These loads were pre-processed before they were inputted on the finite element model, developed using the finite element software ANSYS. The considered conical grout connection, which connects the monopile to the transition piece through friction, was modeled under a changing-status nonlinearity condition. To model the soil–pile interaction, a p-y model was applied using the ANSYS APDL commands. Static, modal, and transient structural analyses were produced to study the structure suitability for its use on offshore environments. Different soil interactions were modeled, and their results were then compared within the transient and modal analysis, indicating that the angle of the grout connection strongly affects the loading conditions on the grout. Moreover, scouring affects the dynamic behavior of the overall supporting structures, thus protection against this phenomenon is suggested.
- 1 June 2019
- Published by Elsevier in Energy Policy
- Vol. 129, 1364-1371
European offshore wind capital cost trends up to 2020
- 1 June 2019
- Published by Elsevier in Energy Policy
- Vol. 129, 1364-1371
European offshore wind capital cost trends up to 2020
- 1 April 2019
- Published by Elsevier in Reliability Engineering & System Safety
- Vol. 184, 228-239
Remaining useful life estimation in aeronautics: Combining data-driven and Kalman filtering
- 1 April 2019
- Published by Elsevier in Reliability Engineering & System Safety
- Vol. 184, 228-239
Remaining useful life estimation in aeronautics: Combining data-driven and Kalman filtering
- 1 March 2019
- Published by Elsevier in International Journal of Industrial Ergonomics
- Vol. 70, 92-106
Push button design requirements and relations to button architecture elements
- 1 March 2019
- Published by Elsevier in International Journal of Industrial Ergonomics
- Vol. 70, 92-106
Push button design requirements and relations to button architecture elements
- 1 February 2019
- Published by Elsevier in Marine Policy
- Vol. 100, 122-131
Path discussion for offshore wind in Portugal up to 2030
- 1 February 2019
- Published by Elsevier in Marine Policy
- Vol. 100, 122-131
Path discussion for offshore wind in Portugal up to 2030
- 1 January 2019
- Published by IGI Global Scientific Publishing
- p. 1-32
Non-Conventional Technologies Selection
Micro-engineering is nowadays a key industrial area with applications is a wide range of products and sectors. The need for a multiplicity of products fostered the development of several processes and combinations of processes in the world of micro-engineering. There are different feasible alternatives to produce the same kind of product. The manufacturing cost is usually the decision factor to select the best alternative among them. But cost is affected by dozens of factors and if not properly modelled causes controversy in so complex decisions. In this chapter, the application of process-based cost modelling is proposed as the engine to identify the best performance spaces for each alternative, using its potential for sensitive analysis of uncertain and/or critical parameters. To illustrate the approach, a case study is developed analyzing four alternatives for the production of a light diffuser with micro-features imbibed, involving micro-injection molding, hot-embossing, micromachining, and powder-injection molding.
- 1 January 2019
- Published by IGI Global Scientific Publishing
- p. 1-32
Non-Conventional Technologies Selection
Micro-engineering is nowadays a key industrial area with applications is a wide range of products and sectors. The need for a multiplicity of products fostered the development of several processes and combinations of processes in the world of micro-engineering. There are different feasible alternatives to produce the same kind of product. The manufacturing cost is usually the decision factor to select the best alternative among them. But cost is affected by dozens of factors and if not properly modelled causes controversy in so complex decisions. In this chapter, the application of process-based cost modelling is proposed as the engine to identify the best performance spaces for each alternative, using its potential for sensitive analysis of uncertain and/or critical parameters. To illustrate the approach, a case study is developed analyzing four alternatives for the production of a light diffuser with micro-features imbibed, involving micro-injection molding, hot-embossing, micromachining, and powder-injection molding.