Abstract
Deep learning models have been applied for varied electrical applications in smart grids with a high degree of reliability and accuracy. The development of deep learning models requires the historical data collected from several electric utilities during the training of the models. The lack of historical data for training and testing of developed models, considering security and privacy policy restrictions, is considered one of the greatest challenges to machine learning-based techniques. The paper proposes the use of homomorphic encryption, which enables the possibility of training the deep learning and classical machine learning models whilst preserving the privacy and security of the data. The proposed methodology is tested for applications of fault identification and localization, and load forecasting in smart grids. The results for fault localization show that the classification accuracy of the proposed privacy-preserving deep learning model while using homomorphic encryption is 97–98%, which is close to 98–99% classification accuracy of the model on plain data. Additionally, for load forecasting application, the results show that RMSE using the homomorphic encryption model is 0.0352 MWh while RMSE without application of encryption in modeling is around 0.0248 MWh.
Highlights
The information layer embedded in the two-way communication electric smart grid has various data sources, such as Advanced Metering Infrastructure (AMI), smart electrical measurement sensors, smart meters, detectors, Phasor Measurement Units (PMU), Remote Terminal Unit (RTU), SupervisoryControl, and Data Acquisition System (SCADA), etc. [1]
The results indicate that the machine learning model without encryption has a coefficient of variation (CV) of 7% and the CV is around 10% when encryption is employed
A privacy-preserving deep learning model that is based on homomorphic encryption is presented for classification and a classical machine learning model based on the Paillier scheme for homomorphic encryption is presented for regression
Summary
The information layer embedded in the two-way communication electric smart grid has various data sources, such as Advanced Metering Infrastructure (AMI), smart electrical measurement sensors, smart meters, detectors, Phasor Measurement Units (PMU), Remote Terminal Unit (RTU), SupervisoryControl, and Data Acquisition System (SCADA), etc. [1]. The collected data represent the behavior of the customers and other network players Those data contain sensitive information about customers, electricity consumption, trading, and operation of electricity distribution networks [3]. Those data can be utilized to develop machine learning algorithms to improve the electrical utility operation in terms of demand response, peak load shaving, fault analysis, etc. There are four main pillars needed to achieve privacy-preserving machine learning; namely, training data privacy, model input privacy, model weights privacy, and model output privacy.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
