Deep Learning-Based Magnetic Coupling Detection for Advanced Induction Heating Appliances

Lucia, O. (Universidad de Zaragoza) ; Navarro, D. (Universidad de Zaragoza) ; Guillen, P. (Universidad de Zaragoza) ; Sarnago, H. (Universidad de Zaragoza) ; Lucia, S.

Resumen: Induction heating has become the reference technology for domestic heating applications due to its benefits in terms of performance, efficiency and safety, among others. In this context, recent design trends aim at providing highly flexible cooking surfaces composed of multi-coil structures. As in many other wireless power transfer systems, one of the main challenges to face is the proper detection of the magnetic coupling with the induction heating load in order to provide improved thermal performance and safe power electronic converter operation. This is specially challenging due to the high variability in the materials used in cookware as well as the random pot placement in flexible induction heating appliances. This paper proposes the use of deep learning techniques in order to provide accurate area overlap estimation regardless of the used pot and its position. An experimental test-bench composed of a complete power converter, multi-coil system and real-Time measurement system has been implemented and used in this study to characterize the parameter variation with overlapped area. Convolutional neural networks are then proposed as an effective method to estimate the covered area, and several implementations are studied and compared according to their computational cost and accuracy. As a conclusion, the presented deep learning-based technique is proposed as an effective tool to estimate the magnetic coupling between the coil and the induction heating load in advanced induction heating appliances.

Idioma: Inglés

DOI: 10.1109/ACCESS.2019.2960109

Año: 2019

Publicado en: IEEE Access 7 (2019), 181668-181677
ISSN: 2169-3536

Factor impacto JCR: 3.745 (2019)

Categ. JCR: COMPUTER SCIENCE, INFORMATION SYSTEMS rank: 35 / 155 = 0.226 (2019) - Q1 - T1
Categ. JCR: ENGINEERING, ELECTRICAL & ELECTRONIC rank: 61 / 265 = 0.23 (2019) - Q1 - T1
Categ. JCR: TELECOMMUNICATIONS rank: 26 / 89 = 0.292 (2019) - Q2 - T1

Factor impacto SCIMAGO: 0.775 - Engineering (miscellaneous) (Q1) - Computer Science (miscellaneous) (Q1) - Materials Science (miscellaneous) (Q2)

Financiación: info:eu-repo/grantAgreement/ES/DGA/FSE
Financiación: info:eu-repo/grantAgreement/ES/MECD-DGA-FSE/FPU17-01442
Financiación: info:eu-repo/grantAgreement/ES/MICINN-AEI/RTC-2017-5965-6
Financiación: info:eu-repo/grantAgreement/ES/MINECO/TEC2016-78358-R

Tipo y forma: Article (Published version)

Área (Departamento): Área Tecnología Electrónica (Dpto. Ingeniería Electrón.Com.)