Francisco Javier de la Calle Herrero

Dr. Francisco Javier de la Calle Herrero

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 September 2016 in Journal of Nondestructive Evaluation
Verified
Journal Article
Closed access
PDF file
FBFrancisco Gonzalez Bulnes
Francisco Gonzalez Bulnes
DGDaniel F. García
Daniel F. García
FHFrancisco Javier de la Calle Herrero
Francisco Javier de la Calle Herrero
Ruben UsamentiagaRuben Usamentiaga
Ruben Usamentiaga
JMJulio Molleda
Julio Molleda

9 February 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Luis Magadán CoboLuis Magadán Cobo
Luis Magadán Cobo
FSFrancisco J. Suárez
Francisco J. Suárez
JGJuan C. Granda
Juan C. Granda
FHFrancisco Javier de la Calle Herrero
Francisco Javier de la Calle Herrero
DGDaniel F. García
Daniel F. García
Remaining useful lifetime (RUL) predictions of electric motors are of vital importance in the maintenance and reduction of repair costs. Thanks to technological advances associated with Industry 4.0, physical models used for prediction and prognostics have been replaced by data-driven models that do not require specialized staff for feature selection, as the model itself learns what features are important. However, these models are usually trained and tested with the same datasets. That makes it difficult to reuse models with different datasets, so they should be retrained with data from the specific motor being analyzed. This paper presents a novel and robust health prognostics technique that predicts the remaining useful lifetime of the bearings of electric motors under different motor conditions (shaft frequency, load, type of bearing) without retraining or fine-tuning the model used. The model integrates the frequency-domain signal analysis and a stacked autoencoder (SAE) with a bidirectional long short-term memory (BiLSTM) neural network. The proposed model is trained with the IMS-bearing dataset and is then tested with IMS, FEMTO, and XJTU-SY datasets without retraining it, providing accurate results in all of them, and proving its robustness with different electric motors and work conditions.

2 May 2018 in IEEE Transactions on Industry Applications
Verified
Journal Article
Closed access
FHFrancisco Javier de la Calle Herrero
Francisco Javier de la Calle Herrero
DGDaniel F. García
Daniel F. García
Ruben UsamentiagaRuben Usamentiaga
Ruben Usamentiaga
In this paper a surface inspection system for rails is proposed. There is a lack of commercial systems and publications about surface inspection of rails. Therefore, the quality control developed by rail manufacturers depends on their own developed systems or on the equipments provided by very few sellers. Commercial systems must be configured manually by experts of seller companies. The configuration process requires large sets of all types of manufactured rails along several periods of time and the active participation of the quality engineers of the manufacturer. This is a long, cumbersome and very expensive process. In this paper we propose a new system that can be configured using a systematic method that can be performed by the quality engineers of the manufacturer. The proposed inspection system uses differential images of the rail surfaces obtained with a technique called the Spectral Images Differentiation Procedure. These images are processed using a computer vision algorithm that looks for variations in the pixel values among the images. In order to offer more information, a neural network approach is used for classifying detected defects into six types. The proposed system is an open solution for the inspection of rail surface, easy to implement at an affordable cost, which can be systematically configured by the quality engineers of the manufacturing company.

16 February 2021 in IEEE Transactions on Industry Applications
Verified
Journal Article
Closed access
FHFrancisco Javier de la Calle Herrero
Francisco Javier de la Calle Herrero
DGDaniel F. García
Daniel F. García
Ruben UsamentiagaRuben Usamentiaga
Ruben Usamentiaga
In this article, a surface inspection system for rails is presented. Rails must meet the strict requirements of international quality standards; however, there are few commercial surface inspection systems for rails and also a lack of publications describing the design and configuration of inspection systems in detail. Therefore, manufacturers must develop their own systems or buy one of the few commercial ones available. These systems also need a long, cumbersome, and expensive configuration process that the manufacturer cannot perform without the assistance of the inspection system provider. The system proposed in this article needs a set of samples and the requirements of the international standards to carry out an automatic configuration process avoiding the cost of manual configuration. The system uses four profilometers to acquire the surface of the rail. The acquired data is compared to a mathematical model of the rail to generate differential topographic images of the surface of the rail. Then a computer vision algorithm is used to detect defects based on the tolerances established in the international quality standards. The system has been tested and validated using a set of rails and a rail pattern from ArcelorMittal, with better results than the other two systems installed in a factory. 1

23 January 2018 in IEEE Transactions on Industry Applications
Verified
Journal Article
Closed access
Ruben UsamentiagaRuben Usamentiaga
Ruben Usamentiaga
DGDaniel F. García
Daniel F. García
FHFrancisco Javier de la Calle Herrero
Francisco Javier de la Calle Herrero
Calibration and registration are two challenging tasks required by a real-time inspection system for long steel products based on multiple three-dimensional sensors. Spatial calibration is required to transform the information acquired by each sensor into a common reference system. Registration is used to align the calculated shape with the model of the product being inspected, which enables the calculation of quality control parameters. This paper proposes efficient solutions for these two problems. Calibration is performed using a specially designed calibration plate that can be used to detect the position and orientation of each sensor. Registration is applied using an analytic alignment between the data and the model. The proposed methods are specially designed to perform accurately in industrial environments where inspection systems are exposed to adverse conditions, such as dust, oil, or water, which provoke noise in the signals. Moreover, vibrations affect the position of the inspected product randomly, which makes accurate measurements much more difficult to achieve. The proposed procedures are validated with extensive tests using real and simulated data showing excellent performance.

16 October 2020 in Applied Optics
Verified
Journal Article
Closed access
Ruben UsamentiagaRuben Usamentiaga
Ruben Usamentiaga
DGDaniel Garcia
Daniel Garcia
FHFrancisco Javier de la Calle Herrero
Francisco Javier de la Calle Herrero
FDFrancisco Delacalle
Francisco Delacalle
Camera calibration requires three steps: estimation of correspondences between world and image coordinates, computation of a linear solution, and nonlinear optimization using the linear estimate as a starting point. The resulting accuracy depends mostly on the first and final steps. However, the nonlinear optimization method can achieve an accurate result only when given an initial estimate close to the global solution. Therefore, obtaining a good linear estimation is crucial for the performance of the camera calibration procedure. This work proposes a robust method to estimate a linear solution for the calibration of line-scan cameras, resulting in individual intrinsic and extrinsic parameters by using only a single line scan. The calculated parameters can then be used by nonlinear optimization methods to finely adjust the estimation of all the line-scan camera parameters, including distortions. The proposed procedure does not impose restrictions on particular orientations, and always generates a well-conditioned problem that can be solved analytically with no optimization required. Extensive experiments are performed to verify the robustness and accuracy of the proposed method. The comparative results demonstrate that the proposed method provides excellent performance.