RT Journal Article
T1 Hybrid Model Development for HVAC System in Transportation
A1 Gálvez, Antonio
A1 Seneviratne, Dammika
A1 Galar, Diego
AB Hybrid models combine physics-based models and data-driven models. This combination is a useful technique to detect fault and predict the current degradation of equipment. This paper proposes a physics-based model, which will be part of a hybrid model, for a heating, ventilation, and air conditioning system installed in the passenger vehicle of a train. The physics-based model is divided into four main parts: heating subsystems, cooling subsystems, ventilation subsystems, and cabin thermal networking subsystems. These subsystems are developed when considering the sensors that are located in the real system, so the model can be linked via the acquired sensor data and virtual sensor data to improve the detectability of failure modes. Thus, the physics-based model can be synchronized with the real system to provide better simulation results. The paper also considers diagnostics and prognostics performance. First, it looks at the current situation of the maintenance strategy for the heating, ventilation, air conditioning system, and the number of failure modes that the maintenance team can detect. Second, it determines the expected improvement using hybrid modelling to maintain the system. This improvement is based on the capabilities of detecting new failure modes. The paper concludes by suggesting the future capabilities of hybrid models.
SN 2227-7080
YR 2021
FD 2021-03
LK https://hdl.handle.net/11556/4577
UL https://hdl.handle.net/11556/4577
LA eng
NO Gálvez , A , Seneviratne , D & Galar , D 2021 , ' Hybrid Model Development for HVAC System in Transportation ' , Technologies , vol. 9 , no. 1 , 18 . https://doi.org/10.3390/technologies9010018
NO Publisher Copyright: © 2021 by the authors.
NO This research was funded by the Basque Government, through ELKARTEK (ref. KK-2020/00049) funding grant.
DS TECNALIA Publications
RD 31 jul 2024