$\bm {P}^{\bm {3}}$: Privacy-Preserving Prediction of Real-Time Energy Demands in EV Charging Networks

Li, Beibei; Guo, Yuqing; Du, Qingyun; Zhu, Ziqing; Li, Xiaohui; Lu, Rongxing

doi:10.1109/tii.2022.3182972

Cited by 15 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The adoption of EV energy consumption models for midemand modeling demonstrates a solution to the scarcity of microscopic EV traffic flow data. This enables the real-time prediction of EV charging requirements and facilitates effective charging coordination based on individual energy requirements, in contrast to real-time charger requirement modeling in [13]- [15]. Nonetheless, the SAR model developed in this work is based on EV traffic data for a single working day, due to the license limitations of the TomTom data collection portal.…”

Section: Discussionmentioning

confidence: 99%

“…These datasets provide information about the starts and end times of charging, the durations of charging sessions, and the amount of energy delivered at predefined charging locations. This approach is adopted in [10], [11] and [12] using published charging datasets collected over long periods of time from different EV chargers, and in [13]- [15] using localized charging data to train machine learning (ML) and deep learning (DL) models for real-time prediction of the EV electricity requirements at specific charging stations. These charging data-based models, however, fail to incorporate the spatial and temporal variations in driving patterns, traveling distances, and driving durations within its prediction model, as they are based on static charging data with no EV mobility considerations.…”

Section: A Demand Modeling Using Charger Datasetsmentioning

confidence: 99%

“…For instance, the data used in [10] is published online at [16] and reflects charging sessions from fast chargers placed in Caltech, California, U.S., and the dataset used in [11] and [12] is also published as part of the open-data ElaadNL project [17]. On the other hand, the data used for developing the prediction models in [13], [14] and [15] are not available online, which hinders the modification and/or generalization of the corresponding models. In addition, since different EV chargers offer varying charging rates and durations [18], it is difficult to generalize patterns of energy drawn from certain chargers to model macroscopic (i.e.…”

Section: A Demand Modeling Using Charger Datasetsmentioning

confidence: 99%

See 2 more Smart Citations

A Data-Driven Approach for EV Electricity Demand Modeling Using Spatial Regression: A UAE Case Study

Elghanam,

Alzaatreh,

Hassan

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The growing global interest in developing environment-friendly and sustainable transportation solutions is motivating mass adoption of Electric Vehicles (EVs). This increasing EV penetration is anticipated to result in a growing electricity demand to address the EV charging requirements. Therefore, precise demand modeling is essential to enable optimal sizing of the electricity generation and distribution networks as well as optimal placement of the EV charging infrastructure. Furthermore, microscopic modeling of EV traffic patterns and trip-wise energy requirements is essential to enable effective charging coordination and demand distribution for on-the-move EVs. However, microscopic EV demand modeling is typically hindered by the scarcity of open-access data that integrates EV charging and driving patterns. Accordingly, this work proposes a methodology for microscopic modeling of the trip-wise electricity demand of mobile EVs in the spatial and temporal domains, using both multiple linear regression and spatial autoregressive models. Secondary open-access data is extracted, wrangled, and pre-processed from a number of data sources to test and validate the proposed methodology on a case study of Dubai -UAE, acknowledging the growing EV adoption rates in the city. The proposed models are benchmarked against baseline models to confirm their superior performance. INDEX TERMSAutoregressive models; demand modeling; electric vehicles; microscopic demand; multiple linear regression; spatial regression.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: A Demand Modeling Using Charger Datasetsmentioning

confidence: 99%

Section: A Demand Modeling Using Charger Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

A Data-Driven Approach for EV Electricity Demand Modeling Using Spatial Regression: A UAE Case Study

Elghanam,

Alzaatreh,

Hassan

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In the seminal paper [48], the Federated Stochastic Gradient Descent (FedSGD) and Federated Averaging (FedAvg) algorithms were first proposed. A recent study [49] employed CNN and bidirectional Long Short-Term Memory (LSTM) networks as local models and implemented FedAvg for global optimization. The study revealed that the performance of the model improved compared to the original local model.…”

Section: Federated Learningmentioning

confidence: 99%

Privacy-Aware Energy Consumption Modeling of Connected Battery Electric Vehicles Using Federated Learning

Yan,

Fang,

et al. 2024

IEEE Trans. Transp. Electrific.

View full text Add to dashboard Cite

Battery Electric Vehicles (BEVs) are increasingly significant in modern cities due to their potential to reduce air pollution. Precise and real-time estimation of energy consumption for them is imperative for effective itinerary planning and optimizing vehicle systems, which can reduce driving range anxiety and decrease energy costs. As public awareness of data privacy increases, adopting approaches that safeguard data privacy in the context of BEV energy consumption modeling is crucial. Federated Learning (FL) is a promising solution mitigating the risk of exposing sensitive information to third parties by allowing local data to remain on devices and only sharing model updates with a central server. Our work investigates the potential of using FL methods, such as FedAvg, and FedPer, to improve BEV energy consumption prediction while maintaining user privacy. We conducted experiments using data from 10 BEVs under simulated real-world driving conditions. Our results demonstrate that the FedAvg-LSTM model achieved a reduction of up to 67.84% in the MAE value of the prediction results. Furthermore, we explored various real-world scenarios and discussed how FL methods can be employed in those cases. Our findings show that FL methods can effectively improve the performance of BEV energy consumption prediction while maintaining user privacy.

show abstract

Legislations and grid codes of vehicle electrification into power grids

Coelho,

Monteiro

2024

Vehicle Electrification in Modern Power Grids

View full text Add to dashboard Cite

$\bm {P}^{\bm {3}}$: Privacy-Preserving Prediction of Real-Time Energy Demands in EV Charging Networks

Cited by 15 publications

References 28 publications

A Data-Driven Approach for EV Electricity Demand Modeling Using Spatial Regression: A UAE Case Study

A Data-Driven Approach for EV Electricity Demand Modeling Using Spatial Regression: A UAE Case Study

Privacy-Aware Energy Consumption Modeling of Connected Battery Electric Vehicles Using Federated Learning

Legislations and grid codes of vehicle electrification into power grids

Contact Info

Product

Resources

About