Elías Herrero Jaraba
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
Elías Herrero Jaraba has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Andrés Llombart Estopiñán
Fundacion CIRCE, Parque Empresarial Dinamiza, Avenida Ranillas 3-D, 1st Floor, 50018 Zaragoza, Spain
3 shared publications - Javier Granado Fornás
CIRCE Technology Center Parque Empresarial Dinamiza, Zaragoza, Spain
3 shared publications David Hernando Polo
2 shared publications- Jesus Martinez del Rincon
Centre for Secure Information Technologies (CSIT), Queen’s University Belfast, Belfast, United Kingdom of Great Britain and Northern Ireland
2 shared publications Julio David Buldain Pérez
2 shared publicationsAnibal Antonio Prada Hurtado
CIRCE Technology Center Parque Empresarial Dinamiza
1 shared publication
Latest Publications
- Open Access
- 5 February 2026
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 168
Hybrid Inception-Transformer model for signals classification: The case of electrical faults in power transformers
- 5 February 2026
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 168
Open AccessHybrid Inception-Transformer model for signals classification: The case of electrical faults in power transformers
- Open Access
- 15 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Hybrid Inception-Transformer model for multivariate time series classification: The case of electrical faults in power transformers
The proposal presented in this paper focuses on the process of classification of electrical faults in a transformer environment. Although the application is very specific in a very concrete field, the scope of the research has a wider scope since it is a multivariate time series classification system. It is based on two already known proposals such as the InceptionTime and Gated Transformer models, in which an attempt is made to extract the benefits provided by each of them. From the first one we extract its optimal computation time, and from the second one the way in which it handles multivariability. Although in the latter case, better results are obtained by modifying and changing the gating proposal for a linear double layer in its output, and the elimination of one of the bottleneck layers in charge of the relationships between the different time variables used. The study focuses on a potential future practical application in which processing time is prioritized together with a high percentage of classification accuracy. Finally, a comparative study is presented between the four models studied or proposed on a proprietary database created for the development of this and future work.
- 15 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessHybrid Inception-Transformer model for multivariate time series classification: The case of electrical faults in power transformers
The proposal presented in this paper focuses on the process of classification of electrical faults in a transformer environment. Although the application is very specific in a very concrete field, the scope of the research has a wider scope since it is a multivariate time series classification system. It is based on two already known proposals such as the InceptionTime and Gated Transformer models, in which an attempt is made to extract the benefits provided by each of them. From the first one we extract its optimal computation time, and from the second one the way in which it handles multivariability. Although in the latter case, better results are obtained by modifying and changing the gating proposal for a linear double layer in its output, and the elimination of one of the bottleneck layers in charge of the relationships between the different time variables used. The study focuses on a potential future practical application in which processing time is prioritized together with a high percentage of classification accuracy. Finally, a comparative study is presented between the four models studied or proposed on a proprietary database created for the development of this and future work.
- Open Access
- 18 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 103431-103451
TransSiamese: A Transformer-Based Siamese Network for Anomaly Detection in Time Series as Approach for Fault Location in Distribution Grids
Fault localization in distribution grids represents a crucial aspect in achieving the concept of smart grids within electrical networks. Despite the existence of various approaches to address this issue, it remains an open and challenging topic in real situations and, therefore, complex grids. One of the main challenges stems from the scarcity of labelled real-world examples is due to the inherent chaotic nature of faults (short circuits between a phase and ground). Obtaining sufficient fault examples for neural network training purposes becomes extremely difficult. Efforts have been made to simulate fault signals and apply data augmentation techniques. However, for fault location specifically, classical augmentation techniques are not applicable due to the unique nature of the fault signals. In this paper, we propose a novel approach to address the problem of extracting fault location information from TDR (Time Domain Reflectometry) signals. This kind of signals, involve injecting pulses into the grid and record these pulses as a bounced signal due to the different impedance changes of the grid. Our approach relies on employing a Transformer for anomaly detection. This Transformer-based Siamese network architecture (abbreviated by TransSiamese) is inspired by TranAD model. This Transformer has been modified to be trained in a Siamese way with a few examples (pre-fault signals/normal state and fault signals/abnormal state). After training phase, we can run inference mode by feeding it signals representing faulty grid states (fault signals). The Transformer attempts to predict the evolution of the signal, however, from a certain point to the end, the predicted fault signal deviates from the True signal supplied. Thanks to the intrinsic property of the Siamese network, it dampens the differences between the learned and the current signal when it comes to pre-fault; on the contrary, it enhances these differences when we are immersed in a fault signal.
- 18 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 103431-103451
Open AccessTransSiamese: A Transformer-Based Siamese Network for Anomaly Detection in Time Series as Approach for Fault Location in Distribution Grids
Fault localization in distribution grids represents a crucial aspect in achieving the concept of smart grids within electrical networks. Despite the existence of various approaches to address this issue, it remains an open and challenging topic in real situations and, therefore, complex grids. One of the main challenges stems from the scarcity of labelled real-world examples is due to the inherent chaotic nature of faults (short circuits between a phase and ground). Obtaining sufficient fault examples for neural network training purposes becomes extremely difficult. Efforts have been made to simulate fault signals and apply data augmentation techniques. However, for fault location specifically, classical augmentation techniques are not applicable due to the unique nature of the fault signals. In this paper, we propose a novel approach to address the problem of extracting fault location information from TDR (Time Domain Reflectometry) signals. This kind of signals, involve injecting pulses into the grid and record these pulses as a bounced signal due to the different impedance changes of the grid. Our approach relies on employing a Transformer for anomaly detection. This Transformer-based Siamese network architecture (abbreviated by TransSiamese) is inspired by TranAD model. This Transformer has been modified to be trained in a Siamese way with a few examples (pre-fault signals/normal state and fault signals/abnormal state). After training phase, we can run inference mode by feeding it signals representing faulty grid states (fault signals). The Transformer attempts to predict the evolution of the signal, however, from a certain point to the end, the predicted fault signal deviates from the True signal supplied. Thanks to the intrinsic property of the Siamese network, it dampens the differences between the learned and the current signal when it comes to pre-fault; on the contrary, it enhances these differences when we are immersed in a fault signal.
- Open Access
- 6 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 23596-23619
Modeling and Simulation of Time Domain Reflectometry Signals on a Real Network for Use in Fault Classification and Location
Today, the classification and location of faults in electrical networks remains a topic of great interest. Faults are a major issue, mainly due to the time spent to detect, locate, and repair the cause of the fault. To reduce time and associated costs, automatic fault classification and location is gaining great interest. State-of-the-art techniques to classify and locate faults are mainly based on line-impedance measurements or the detection of the traveling wave produced by the event caused by the fault itself. In contrast, this paper describes the methodology for creating a database and a model for a complex distribution network. Both objectives are covered under the paradigm of the time-domain pulse reflectometry (TDR) principle. By using this technique, large distances can be monitored on a line with a single device. Thus, in this way a database is shared and created from the results of simulations of a real and complex distribution network modeled in the PSCADTM software, which have been validated with measurements from an experimental test setup. Experimental validations have shown that the combination of the TDR technique with the modeling of a real network (including the real injector and the network coupling filter from the prototype) provides high-quality signals that are very similar and reliable to the real ones. In this sense, it is intended firstly that this model and its corresponding data will serve as a basis for further processing by any of the existing state-of-the-art techniques. And secondly, to become a valid alternative to the already well-known Test Feeders but adapted to work groups not used to the electrical world but to the environment of pure data processing.
- 6 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 23596-23619
Open AccessModeling and Simulation of Time Domain Reflectometry Signals on a Real Network for Use in Fault Classification and Location
Today, the classification and location of faults in electrical networks remains a topic of great interest. Faults are a major issue, mainly due to the time spent to detect, locate, and repair the cause of the fault. To reduce time and associated costs, automatic fault classification and location is gaining great interest. State-of-the-art techniques to classify and locate faults are mainly based on line-impedance measurements or the detection of the traveling wave produced by the event caused by the fault itself. In contrast, this paper describes the methodology for creating a database and a model for a complex distribution network. Both objectives are covered under the paradigm of the time-domain pulse reflectometry (TDR) principle. By using this technique, large distances can be monitored on a line with a single device. Thus, in this way a database is shared and created from the results of simulations of a real and complex distribution network modeled in the PSCADTM software, which have been validated with measurements from an experimental test setup. Experimental validations have shown that the combination of the TDR technique with the modeling of a real network (including the real injector and the network coupling filter from the prototype) provides high-quality signals that are very similar and reliable to the real ones. In this sense, it is intended firstly that this model and its corresponding data will serve as a basis for further processing by any of the existing state-of-the-art techniques. And secondly, to become a valid alternative to the already well-known Test Feeders but adapted to work groups not used to the electrical world but to the environment of pure data processing.
- Open Access
- 14 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 110521-110536
Detection and Classification of Fault Types in Distribution Lines by Applying Contrastive Learning to GAN Encoded Time-Series of Pulse Reflectometry Signals
This study proposes a new method for detecting and classifying faults in distribution lines. The physical principle of classification is based on time-domain pulse reflectometry (TDR). These high-frequency pulses are injected into the line, propagate through all of its bifurcations, and are reflected back to the injection point. According to the impedances encountered along the way, these signals carry information regarding the state of the line. In the present work, an initial signal database was obtained using the TDR technique, simulating a real distribution line using (PSCAD™). By transforming these signals into images and reducing their dimensionality, these signals are processed using convolutional neural networks (CNN). In particular, in this study, contrastive learning in Siamese networks was used for the classification of different types of faults (ToF). In addition, to avoid the problem of overfitting owing to the scarcity of examples, generative adversarial neural networks (GAN) have been used to synthesise new examples, enlarging the initial database. The combination of Siamese neural networks and GAN allows the classification of this type of signal using only synthesised examples to train and validate and only the original examples to test the network. This solves the problem of the lack of original examples in this type of signal of natural phenomena which are difficult to obtain and simulate.
- 14 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 110521-110536
Open AccessDetection and Classification of Fault Types in Distribution Lines by Applying Contrastive Learning to GAN Encoded Time-Series of Pulse Reflectometry Signals
This study proposes a new method for detecting and classifying faults in distribution lines. The physical principle of classification is based on time-domain pulse reflectometry (TDR). These high-frequency pulses are injected into the line, propagate through all of its bifurcations, and are reflected back to the injection point. According to the impedances encountered along the way, these signals carry information regarding the state of the line. In the present work, an initial signal database was obtained using the TDR technique, simulating a real distribution line using (PSCAD™). By transforming these signals into images and reducing their dimensionality, these signals are processed using convolutional neural networks (CNN). In particular, in this study, contrastive learning in Siamese networks was used for the classification of different types of faults (ToF). In addition, to avoid the problem of overfitting owing to the scarcity of examples, generative adversarial neural networks (GAN) have been used to synthesise new examples, enlarging the initial database. The combination of Siamese neural networks and GAN allows the classification of this type of signal using only synthesised examples to train and validate and only the original examples to test the network. This solves the problem of the lack of original examples in this type of signal of natural phenomena which are difficult to obtain and simulate.
- 3 February 2011
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 7 (4) , 267-279
Real-time GPU color-based segmentation of football players
- 3 February 2011
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 7 (4) , 267-279
Real-time GPU color-based segmentation of football players
- 16 December 2009
- Published by Springer Nature in Journal of Mathematical Imaging and Vision
- Vol. 36 (3) , 241-253
Gaussian Approximation for Tracking Occluding and Interacting Targets
- 16 December 2009
- Published by Springer Nature in Journal of Mathematical Imaging and Vision
- Vol. 36 (3) , 241-253
Gaussian Approximation for Tracking Occluding and Interacting Targets
- 1 January 2009
- Published by Springer Nature
- p. 127-138
Calibration and Comparison of Visual Tracking on the Ground for Multi Camera Tracking
- 1 January 2009
- Published by Springer Nature
- p. 127-138
Calibration and Comparison of Visual Tracking on the Ground for Multi Camera Tracking