Browse Prior Art Database

Privacy-Preserving State Estimation and Prediction with Electricity Data Disclosure Number: IPCOM000244732D
Publication Date: 2016-Jan-06
Document File: 8 page(s) / 210K

Publishing Venue

The Prior Art Database

Related People

Matus Harvan: INVENTOR [+5]


The invention enables the computation of the current state and predictions without disclosing the actual input data based on a cryptographic technique called homomorphic encryption. A DSO or another service provider can provide the applications listed in Section 1 without gaining access to the measurement data from each customer. There are three dinstinct novel features: The first enables privacy-preserving state estimation by aggregating the cumulative state from the current state of individual households. The privacy of the home owners' data is preserved in that the DSO and aggregators between smart meters and DSO only learn about the entire state and not about any individual contributions. The second feature enables the participation in energy trading through the information about future power consumption and generation in the DSO's virtual power plant(s). To this end, the home owners provide information about their own planned/scheduled power consumption and generation to the DSO, again in a way that enables the DSO to learn about the consumption and generation in the entire virtual power plant but not about individual plans. The third feature is aggregation functionality beyond additive functions (sums and (weighted) averages) without the overhead of leveled or fully homomorphic encryption. The solutions and optimizations developed as part of the SecreDS projects ensure that the computational demands remain feasible and realistic.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 15% of the total text.

Page 01 of 8

Dokumentenart/Type of document

Klassifizierung/Classification Ablageinformation (Reg.-Nr/Dateiname)/Inform. for filing

Ort/Pl. von Abt./from Dept verfasst von Hr./Fr./Frl./from Mr./Mrs./Miss Tel. int./Phone

Invention Disclosure

Erstellungsdatum: Seiten/Pages




Matus Harvan

Thomas Locher Martin Naef Sebastian Obermeier Yvonne Anne Pignolet

Johannes Schneider



Invention Disclosure

Ort/Pl. an Abt./to Dept.

an Hr./Fr./Frl./to Mr./Mrs./Miss

Nr. Bemerkungen/Notes

Privacy-Preserving State Estimation and Prediction with Electricity Data

A number of innovations promised by "Smart Grids" rely on high-resolution metering data from the electricity consumers - typically provided by "smart meters". Such application include:

• Distribution system or feeder automation that relies on improved observability of the low- voltage part of the network. Specific applications include the control of solar inverters, distributed energy resources (e.g., batteries), and other active elements (e.g., tap changers) to guarantee voltage stability in the network.

• Adaptive protection schemes that can handle current flows not visible at the secondary substation or ring main unit - e.g., between distributed energy resources and consumers attached to the same feeder branch.

• Aggregation of small consumers and producers to participate in the energy market. This includes balance group optimization or demand side management, appearing in the form of a virtual power plant (VPP).

All these applications require real-time or near-real-time measurements from the meters to compute aggregate values (e.g., sum of currents at a specific node in the electric network, the total controllable power consumption in a VPP).

2.Problem Description

The installation of smart meters operating at high resolution is received with significant concerns about privacy. Access to the measurement data enables a detailed insight into the habits and lifestyle of consumers. The following paragraph from Wikipedia summarizes the issue (

CH-1516401_20160106_PP_01.docx 1 / 8

Page 02 of 8

"One technical reason for privacy concerns is that these meters send detailed information

about how much electricity is being used each time. More frequent reports provide more detailed information. Infrequent reports may be of little benefit for the provider, as it doesn't allow as good demand management in the response of changing needs for electricity. On the other hand, very frequent reports would allow to the utility company to infer behavioral patterns for the occupants of a house, such as when the members of the household are probably asleep or absent. Current trends are to increase the frequency of reports. A solution which benefits both the provider and the user's privacy, would be to adapt the interval dynamically. In BC Canada the electric utility is government owned and as such must c...