Di-PriDA: Differentially Private Distributed Load Balancing Control for the Smart Grid

Liao, Xiaojing; Srinivasan, Preethi; Formby, David; Beyah, Raheem

doi:10.1109/tdsc.2017.2717826

Cited by 28 publications

(13 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A myriad of works [Eibl and Engel, 2017,Ács and Castelluccia, 2011, Zhao et al, 2014, Liao et al, 2017, Barbosa et al, 2016, Mak et al, 2019, Yang et al, 2017 studied the effect of DP on aggregated energy data in protecting privacy of users while collecting the fine-grained meter readings. Another privacy-preserving protocol using DP was presented in [Danezis et al, 2011] by adding noise to the aggregate energy consumption (bills) of consumers to hide their activities that might be revealed from their energy consumption traces.…”

Section: Related Workmentioning

confidence: 99%

Do Auto-Regressive Models Protect Privacy? Inferring Fine-Grained Energy Consumption From Aggregated Model Parameters

Sheikh

Asghar

Farokhi

et al. 2022

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

We investigate the extent to which statistical predictive models leak information about their training data. More specifically, based on the use case of household (electrical) energy consumption, we evaluate whether white-box access to auto-regressive (AR) models trained on such data together with background information, such as household energy data aggregates (e.g., monthly billing information) and publicly-available weather data, can lead to inferring fine-grained energy data of any particular household. We construct two adversarial models aiming to infer fine-grained energy consumption patterns. Both threat models use monthly billing information of target households. The second adversary has access to the AR model for a cluster of households containing the target household. Using two real-world energy datasets, we demonstrate that this adversary can apply maximum a posteriori estimation to reconstruct daily consumption of target households with significantly lower error than the first adversary, which serves as a baseline. Such fine-grained data can essentially expose private information, such as occupancy levels. Finally, we use differential privacy (DP) to alleviate the privacy concerns of the adversary in dis-aggregating energy data. Our evaluations show that differentially private model parameters offer strong privacy protection against the adversary with moderate utility, captured in terms of model fitness to the cluster.Keywords aggregate statistics • white-box attacks • inference attacks • auto-regressive models • differential privacy • energy data privacy.

show abstract

Section: Related Workmentioning

confidence: 99%

Do Auto-Regressive Models Protect Privacy? Inferring Fine-Grained Energy Consumption From Aggregated Model Parameters

Sheikh

Asghar

Farokhi

et al. 2022

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

show abstract

“…To guarantee electricity services and applications while protecting customer privacy, lots of privacy preserving approaches have been proposed, which can be classified into two categories: cryptography-based and non-cryptographybased privacy preserving schemes (Liao et al, 2019).…”

Section: Privacymentioning

confidence: 99%

Smarter Grid in the 5G Era: A Framework Integrating Power Internet of Things With a Cyber Physical System

et al. 2021

View full text Add to dashboard Cite

As the energy infrastructure of smart cities, smart grid upgrades traditional power grid systems with state-of-the-art information and communication technologies. In particular, as the full deployment of the Internet of Things in the power grid (a.k.a. power Internet of Things or PIoT), the newly introduced information flow together with inherent energy flow makes it more efficient for power generation, transmission, distribution, and consumption. To further exploit the precious energy and the latest 5G technologies, this article boosts to add a value flow in the smart grid, mainly including the value created by innovative services and market mechanisms and the value added by the information flow. Specifically, by integrating PIoT with cyber-physical systems, this article sketches a conceptual framework of the cyber-physical power system (CPPS). The CPPS carries out holistic perception and ubiquitous connection of distributed energy sources and electrical facilities and builds up a smarter power grid with global information interaction, intelligent decision-making, and real-time agile control. Finally, for illustration purposes, we conduct a case study regarding an intelligent home management system.

show abstract

“…6 (under the condition ε=8, δ=10 −5 ). In each communication round, only 30% of clients (e.g., 15 clients when = 50) are selected to participate in the training process. Unlike feeder-level load forecasting, which has a regular peak load every day, household-level load forecasting is more challenging as the load profile in different days vary a lot.…”

Section: E Comparison Of the Proposed Model With Other Algorithmsmentioning

confidence: 99%

“…The FL-based system can train a reinforcement model that can better manage the distributed generation, energy storage system and appliance usages inside smart homes. Moreover, privacy-preserving appliance load scheduling methods are proposed based on differential privacy [15] and MPC [16], respectively.…”

Section: Introductionmentioning

confidence: 99%

Differential Private Federated Learning for Privacy-Preserving Third Party Service Framework in Advanced Metering Infrastructure

Zhang¹,

Cordoba-Pachon²,

Watkins³

et al. 2021

Preprint

View full text Add to dashboard Cite

The advanced metering infrastructure (AMI) is a compulsory component of the future smart grid; it not only provides near real-time two-way communication between the consumers and the utility but also gives an opportunity to third parties to provide relevant value-added services to the consumers to improve the user satisfaction. However, existing services require the consumers share their private energy data with other parties, which has potential privacy risks. To better balance the excellent quality of the services and privacy guarantee, a novel differential private federated learning-based third-party service platform is proposed. Instead of sending the original energy data to the cloud server, the central server in the proposed scheme only collects the model parameters, which are trained locally inside the consumers’ houses. Then the collected parameters are aggregated differential privately to eliminate the identity of individuals, and the aggregated parameters are used to update the central model and improve the model performance. Furthermore, a novel attention-based bidirectional long short-term memory neural network model is adopted to make predictions. In the case study, a residential short term load forecasting task is implemented to evaluate the performance of the proposed model; from the simulation results, the conclusion is made that the proposed model can achieve similar accuracy as the typical centralized model and better control the privacy loss flexibly at the same time. <br>

show abstract

Di-PriDA: Differentially Private Distributed Load Balancing Control for the Smart Grid

Cited by 28 publications

References 25 publications

Do Auto-Regressive Models Protect Privacy? Inferring Fine-Grained Energy Consumption From Aggregated Model Parameters

Do Auto-Regressive Models Protect Privacy? Inferring Fine-Grained Energy Consumption From Aggregated Model Parameters

Smarter Grid in the 5G Era: A Framework Integrating Power Internet of Things With a Cyber Physical System

Differential Private Federated Learning for Privacy-Preserving Third Party Service Framework in Advanced Metering Infrastructure

Contact Info

Product

Resources

About