This study presents the experimental results of the effects on base oils after thermal and electrical aging to determine key parameters of next-generation fluids for thermal management in electric vehicles. The test fluids selected were a mineral base oil API G-III, an API G-IV Polyalphaolefin (PAO), a diester, and a polyolester, all of which had similar kinematic viscosity (KV100 = 4 cSt). All were initially characterized with measurements of density, viscosity, thermal conductivity, specific heat capacity, breakdown voltage, resistivity, and dissipation factor. They underwent two separate aging processes, one thermal, heating the test fluid at 150 °C for 120 h with a copper strip as a catalyst; and the second one an electrical aging process, with the application of 1000 breakdown voltage discharges. The same properties were measured again after each aging process and compared to the initial ones. It was found that the thermal properties ranged with similar values and did not suffer major changes after the aging processes, unlike electrical properties, which vary between samples and after thermal and electrical stress. The insights gained from this study have implications for both the development of next-generation e-thermal fluids and the future standardization of these fluids for EV thermal management applications. The findings of this study underscore the significance of formulating and selecting a suitable dielectric fluid for EV thermal management. By leveraging the insights provided, researchers and engineers can advance in the development of efficient and reliable e-thermal fluids while working towards future standardization to enhance the performance and safety of EV battery systems.
This study aims to assess the compatibility of various e-thermal fluids for immersion cooling in battery electric vehicles through a copper wire resistance corrosion test. The tested fluids include a polyalphaolefin, diester, mineral oil API G-III, transformer oil, and a fully formulated dielectric coolant. The test was conducted at 130 °C for 336 h, and the resistance of the copper wires was monitored in vapor and oil phases. By comparing the resistance variation and analyzing portions of the wires through scanning electron microscopy, it was found that the vapor phase of PAO and diester in one of the tests exhibited significant corrosion, while the dielectric coolant showed minimal corrosive effects, implying better compatibility. These results provide insights into the corrosion behavior and compatibility of the fluids with copper, which are essential for selecting suitable dielectric fluids for immersion cooling applications in electric vehicles.
The adoption of predictive maintenance in public transportation has gained increasing attention in the context of Industry 4.0. However, many urban bus fleets remain in early digital transformation stages, with limited historical data and fragmented infrastructures that hinder the implementation of data-driven strategies. This study proposes a reproducible Machine Learning pipeline tailored to such data-scarce conditions, integrating domain-informed feature engineering, lightweight and interpretable models (Linear Regression, Ridge Regression, Decision Trees, KNN), SMOGN for imbalance handling, and Leave-One-Out Cross-Validation for robust evaluation. A scheduled batch retraining strategy is incorporated to adapt the model as new data becomes available. The pipeline is validated using real-world data from hybrid diesel buses, focusing on the prediction of time spent in critical soot accumulation zones of the Diesel Particulate Filter (DPF). In Zone 4, the model continued to outperform the baseline during the production test, indicating its validity for an additional operational period. In contrast, model performance in Zone 3 deteriorated over time, triggering retraining. These results confirm the pipeline’s ability to detect performance drift and support predictive maintenance decisions under evolving operational constraints. The proposed framework offers a scalable solution for digitally emerging fleets.
Although urban bus fleets increasingly generate HVAC operational data through onboard sensors and telematics platforms, their practical use for maintenance decision-making remains limited. In real-world conditions, these data are rarely accompanied by reliable fault labels, complete contextual information, or a clear physical interpretation of the monitored variables. As a result, there is still a need for methodological frameworks capable of transforming imperfect operational data into physically meaningful and maintenance-oriented insights. This work proposes a structured framework that combines domain knowledge, unsupervised anomaly detection, and explainable artificial intelligence to support the initial exploitation of HVAC fleet data. The framework begins with sensor contextualization and physical mapping of available measurements onto the vapor-compression refrigeration cycle, enabling the construction of physically meaningful features and a coherent operational baseline. An Isolation Forest model is then used to identify deviations from this baseline, while Kernel SHAP explains the resulting deviation scores. SHAP contributions are further grouped into physics-informed explanation families, allowing anomalies to be interpreted in terms of HVAC mechanisms rather than isolated variables. The results show that this domain knowledge–XAI approach helps distinguish physically plausible rare operating states, inspection candidates, and ambiguous multivariate cases without relying on confirmed fault labels. The main contribution of this study is a physics-informed XAI framework that combines domain knowledge, unsupervised anomaly detection, and explainability methods to transform imperfect HVAC operational data from early-stage digitalized bus fleets into interpretable and maintenance-relevant information.
About Me
pHd candidate in CMT- Motores Térmicos as Universitat Politècnica de València
pHd candidate in CMT- Motores Térmicos as Universitat Politècnica de València