Differentially Private <i>K</i>-Means Clustering Applied to Meter Data Analysis and Synthesis

Ravi, Nikhil; Scaglione, Anna; Kadam, S. D.; Gentz, Reinhard; Peisert, Sean; Lunghino, Brent; Levijarvi, Emmanuel; Shumavon, Aram

doi:10.1109/tsg.2022.3184252

Cited by 12 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They do not fit a Gaussian multivariate distribution directly. In prior work [34], however, we showed that they fit well a multivariate log-Normal distribution, which implies that the logarithm of the daily pattern is a multivariate Gaussian vector, which is what we need to apply our method. Let p[k] be the power consumption at hour k during the day.…”

Section: B Application To Power Measurements Datamentioning

confidence: 86%

Differential Privacy for Class-Based Data: A Practical Gaussian Mechanism

Ramakrishna

Scaglione

et al. 2023

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

In this paper, we present a notion of differential privacy (DP) for data that comes from different classes. Here, the class-membership is private information that needs to be protected. The proposed method is an output perturbation mechanism that adds noise to the release of query response such that the analyst is unable to infer the underlying class-label. The proposed DP method is capable of not only protecting the privacy of class-based data but also meets quality metrics of accuracy and is computationally efficient and practical. We illustrate the efficacy of the proposed method empirically while outperforming the baseline additive Gaussian noise mechanism. We also examine a real-world application and apply the proposed DP method to the autoregression and moving average (ARMA) forecasting method, protecting the privacy of the underlying data source. Case studies on the real-world advanced metering infrastructure (AMI) measurements of household power consumption validate the excellent performance of the proposed DP method while also satisfying the accuracy of forecasted power consumption measurements.

show abstract

Section: B Application To Power Measurements Datamentioning

confidence: 86%

Differential Privacy for Class-Based Data: A Practical Gaussian Mechanism

Ramakrishna

Scaglione

et al. 2023

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

show abstract

“…Utilizing differential privacy in the smart grid has been also studied in the literature [14]- [16]. It was shown that differential privacy is able to prevent load monitoring in smart grids [15].…”

Section: Related Workmentioning

confidence: 99%

LPPDA: A Light-Weight Privacy-Preserving Data Aggregation Protocol for Smart Grids

Kamal,

Al-Ali,

Al-Ali

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Smart meters are continuously being deployed in several countries as a step in the direction of modernizing the power grid. Smart meters allow for automatic electricity consumption reporting to energy providers to facilitate billing and demand-based power generation. However, research has shown that such high resolution reporting to suppliers can potentially be used to invade consumers' privacy; by identifying and predicting their behavior based on their consumption readings. This work presents a new protocol to preserve users' privacy while maintaining the benefits of smart grids. The proposed method utilizes different techniques like randomization, masking, and differential privacy to build the scheme. The proposed method is shown to be more efficient compared to previous work in terms of performance and communication overhead. The implementation, simulation, and analysis are performed on datasets of real smart meters readings of households and electric vehicle chargers.

show abstract

“…However, its ultimate division outcomes hinge on the chosen affiliation threshold, making it challenging to accurately ascertain the peak and valley time interval's demarcation points (DING et al, 2001a;XING et al, 2007;Chong, 2019). The method of cluster analysis involves categorizing data into various groups, each containing similar elements, by analyzing the inherent correlation among data (Ravi et al, 2022). This technique is prevalent in the segmentation of TOU electricity pricing periods due to its resistance to subjective biases and its ability to thoroughly explore correlations across time intervals (QIAO, 2011;DONG and LIN, 2019;JIANG et al, 2021;Lei et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Optimization method of time-of-use electricity price for the cost savings of power grid investment

Dengfeng,

Zhonghao,

Xiaoju

et al. 2024

Front. Energy Res.

View full text Add to dashboard Cite

The concept of time-of-use (TOU) electricity pricing is widely recognized as a key strategy to bridge the gap between electricity availability and consumption, enhance the efficiency of electricity, and refine the patterns of electricity usage. Nonetheless, the existing policy on pricing electricity based on TOU electricity pricing is missing a theoretical approach that evaluates the load properties and the advantages of investing in the power grid. Consequently, the article suggests a method for optimizing electricity prices based on TOU electricity pricing to reduce the costs associated with investing in power grids. Initially, a model for optimizing electricity prices based on TOU electricity pricing is developed, offering support for the pricing strategy of the power grid; Subsequently, a method for dividing TOU electricity pricing using the Gaussian Mixture Module (GMM) clustering algorithm is introduced, offering theoretical backing for the creation of such pricing strategies; Following this, a detailed optimization approach for electricity pricing of electricity TOU electricity pricing is suggested, along with taking into account the benefits of grid investments and the power grid’s load properties, the formulation of the electricity pricing strategy for TOU electricity pricing; Ultimately, this approach is corroborated by the Chongqing power system in China, aiming to minimize disparities in peak load valleys and enhance the advantages of grid investments, thereby offering technical assistance for the scientific determination of TOU electricity pricing.

show abstract

Differentially Private K-Means Clustering Applied to Meter Data Analysis and Synthesis

Cited by 12 publications

References 30 publications

Differential Privacy for Class-Based Data: A Practical Gaussian Mechanism

Differential Privacy for Class-Based Data: A Practical Gaussian Mechanism

LPPDA: A Light-Weight Privacy-Preserving Data Aggregation Protocol for Smart Grids

Optimization method of time-of-use electricity price for the cost savings of power grid investment

Contact Info

Product

Resources

About