Roberto Ferrero

Roberto Ferrero

Auto-generated profile, based on publicly available information

Roberto Ferrero

Department of Electrical Engineering and Electronics, University of Liverpool, Liverpool, UK

Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

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

Representative Publications

View All Publications
13 February 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
XCXiaohan Chen
Xiaohan Chen
Rui YangRui Yang
Rui Yang
YXYihao Xue
Yihao Xue
MHMengjie Huang
Mengjie Huang
RFRoberto Ferrero
Roberto Ferrero
ZWZidong Wang
Zidong Wang
The traditional deep learning-based bearing fault diagnosis approaches assume that the training and test data follow the same distribution. This assumption, however, is not always true for the bearing data collected in practical scenarios, leading to significant decline to fault diagnosis performance. In order to satisfy this assumption, the transfer learning concept is introduced in deep learning by transferring the knowledge learned from other data or models. Due to the excellent capability of feature learning and domain transfer, deep transfer learning methods have gained widespread attention in bearing fault diagnosis in recent years. This article presents a comprehensive review of the development of deep transfer learning-based bearing fault diagnosis approaches since 2016. In this review, a novel taxonomy of deep transfer learning-based bearing fault diagnosis methods is proposed from the perspective of target domain data properties divided by labels, machines, and faults. By covering the whole life cycle of deep transfer learning-based fault diagnosis and discussing the research challenges and opportunities, this review provides a systematic guideline for researchers and practitioners to efficiently identify suitable deep transfer learning models based on the actual problems encountered in bearing fault diagnosis.

7 May 2014 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Giovanni DotelliGiovanni Dotelli
Giovanni Dotelli
RFRoberto Ferrero
Roberto Ferrero
Paola Gallo StampinoPaola Gallo Stampino
Paola Gallo Stampino
Saverio LatorrataSaverio Latorrata
Saverio Latorrata
Sergio ToscaniSergio Toscani
Sergio Toscani
This paper discusses the possibility to use the current ripple introduced by switch mode power converters for low-cost monitoring of polymer electrolyte membrane (PEM) fuel cell (FC) state-of-health, suitable for commercial applications that cannot afford dedicated instrumentation. In more details, an estimate of the ohmic resistance, which is a good indicator of the membrane water content, can be obtained from the high-frequency ripple response by data processing in the frequency domain, while lower frequency ripple at 100/120 Hz (when present) is in the typical frequency range of activation processes. All the available impedance estimates, together with the dc voltage measurement, can be used to promptly detect FC drying and flooding, that are the two opposite failure modes as far as water balance is concerned. The proposed diagnostic approach is tested on a single PEM FC in drying and flooding conditions, by emulating three-phase and single-phase inverter ripples by means of an electronic load.

19 March 2015 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Giovanni DotelliGiovanni Dotelli
Giovanni Dotelli
RFRoberto Ferrero
Roberto Ferrero
Paola Gallo StampinoPaola Gallo Stampino
Paola Gallo Stampino
Saverio LatorrataSaverio Latorrata
Saverio Latorrata
Sergio ToscaniSergio Toscani
Sergio Toscani
In this paper, a low-cost approach for polymer electrolyte membrane fuel cell (FC) drying and flooding diagnosis based on power converter ripple is presented, suitable to be implemented in commercial applications. If proper signal processing algorithms are employed, the high-frequency ripple inherently produced by switch-mode converters allows one to monitor the FC ohmic resistance, which is a good indicator for membrane drying, while an ad hoc switching control of the converter allows one to monitor also the low-frequency impedance, which is an indicator for cell flooding. This technique is tested with a dc/dc boost converter directly connected to a single cell, discussing how different FC operation modes (such as constant current or constant voltage) may affect the sensitivity required to the diagnostic algorithm to correctly recognize drying and flooding occurrences.

20 April 2016 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Giovanni DotelliGiovanni Dotelli
Giovanni Dotelli
RFRoberto Ferrero
Roberto Ferrero
Paola Gallo StampinoPaola Gallo Stampino
Paola Gallo Stampino
Saverio LatorrataSaverio Latorrata
Saverio Latorrata
STSergio Toscani
Sergio Toscani
Polymer electrolyte membrane fuel cell (PEM FC) operation is likely to be characterized by voltage dips on timescales shorter than 1 s, arising from temporary flooding of gas channels or porous layers, particularly when the FC is operated at high humidity levels. If supercapacitors are employed in hybrid systems, they can make up for the temporary lack of energy produced by the FC. However, the steep slopes of the voltage dips affect the energy that can be actually delivered by the supercapacitor because of its series impedance, and this should be taken into account when sizing it. This paper presents a simplified approach for sizing the supercapacitor, based on some observed peculiar features of the FC dips, which allow a simple but accurate model for the evaluation of the supercapacitor response to such dips. The validity of such an approach is supported by simulation and experimental results performed on a single PEM FC and on a supercapacitor.

1 September 2012 in International Journal of Hydrogen Energy
Verified
Journal Article
Closed access
RFR. Ferrero
R. Ferrero
MMMirko Marracci
Mirko Marracci
MPM. Prioli
M. Prioli
BTB. Tellini
B. Tellini

28 December 2012 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
RFRoberto Ferrero
Roberto Ferrero
MMMirko Marracci
Mirko Marracci
BTBernardo Tellini
Bernardo Tellini
In this paper, the results of an experimental analysis performed on a polymer electrolyte membrane fuel cell (FC) for the evaluation of the current ripple effects are reported. The polarization curve (PC) and the impedance spectrum measured for a single FC with a Nafion membrane are presented discussing some advantages and disadvantages of the two adopted techniques from the measurement point of view. The ohmic and activation resistance values provided by the PC are compared to the frequency limits of the equivalent resistance estimated by the spectroscopic analysis to evaluate the additional losses and the system efficiency reduction introduced by the current ripple. The presence of a voltage–current hysteretic behavior is discussed throughout the paper, and some comments on how such a relationship can impact on the identification of the indicated parameters are included.