Privacy vs. NILM: Obfuscating your Power Consumption with Load Hiding

Load-based load hiding approach

With the development and introduction of smart metering, the energy information from costumers changes from infrequent manual meter readings to fine-grained energy consumption data. On the one hand, these measurements will lead to an improvement in costumers’ energy habits, but on the other hand, the fine-grained data produces information about a household and households’ inhabitants, which give rise to privacy issues because these monitoring results disclose user behavior which could be extracted by smart algorithms and techniques. The loss of privacy by load disaggregation and data mining is a huge upcoming smart grid and social issue which enforces the need for privacy-preserving techniques, which can be divided into the following three possibilities:

1. Anonymization of metering data: The metering data and customer identity are separated by a third-party id
2. Privacy-preserving metering data aggregation: Metering data is geographically encapsulated by aggregating the metering data of co-located consumers 
3. Masking and obfuscation of metering data: Masking the power demand by adding or withdrawing the to the meter visible energy demand with the help of rechargeable batteries or controllable loads.

In the paper

D. Egarter, C. Prokop, and W. Elmenreich. Load hiding of household's power demand. In Proc. IEEE International Conference on Smart Grid Communications (SmartGridComm'14), Venice, Italy, 2014.

a state-of-the-art battery-based load hiding (BLH) technique, which uses a controllable battery to disguise the power consumption and a novel load hiding technique called load-based load hiding (LLH) are presented and compared. A load-based load hiding system controls appliances in a specific way to obfuscate a household’s power demand. For example, an electric water boiler could be instrumented to consume energy in a way that masks the power consumption of smaller household devices like coffee machines or a TV. There is no comfort loss expected for the customer: Overall, the boiler will consume a typical amount of energy and produce the expected amount of hot water.
Using this approach, however, reduces the predictability of your energy consumption, which is good for privacy, but a disadvantage for grid operators.

Meetup on Future Research Challenges in Energy Informatics

18 - 19 April 2018 @ University of Klagenfurt, Austria

Objective

This two-day event aims to bring together researchers that are working on the topic of energy informatics in academia. 

The focus of this meetup will be on Non-Intrusive Load Monitoring (NILM). Other relevant subtopics of energy informatics such as data analytics, energy management systems, or artificial intelligence in Smart Microgrids are warmly welcome.

Within a small group, participants will present their latest findings, share experiences, discuss current issues, and discover possible ways of future cooperation and collaboration. 

Researchers interested in joining are asked to apply with a talk title and abstract. Application deadline is March 19th. A committee will decide upon acceptance till March 22nd.

Apply now

Remarks

The organisers would like to highlight that the University of Klagenfurt cannot provide any funding for expenses such as travelling or accommodation. Students of the University of Klagenfurt are eligible to attend the Keynote and Session 1.

Find further information here.

Correlation Filters for Load Classification @ IEEE SmartGridComm

We are happy to announce that our paper "On the Applicability of Correlation Filters for Appliance Detection in Smart Meter Readings" was accepted and presented at this year's SmartGridComm conference in Dresden.

With our load classification approach based on correlation filters, we aim to provide a low-cost non-Intrusive Load Monitoring (NILM) method for measurement equipment with limited computational capabilities such as networked sensors or smart plugs. One of these small devices to run such an algorithm on would be our YoMo metering board.

Abstract:

"Communication systems utilise correlation filters to detect waveforms. In a broader sense, these filters examine the amount of resemblance between a template pattern and the input pattern. In the domain of smart grids, many applications require the detection of active electrical appliances, their condition as well as their current state of operation. Furthermore, the identification of power eaters, the recognition of ageing effects, and the forecast of required maintenance represent important challenges in (home) energy management systems.

In this paper, we examine the applicability of correlation filters as a possible solution to meet such challenges. First, we introduce the concept of predictability to power consumption patterns of electrical appliances. Second, we present our concept and the implementation of correlation filters for this kind of application. The correlation filters utilise a particular consumption pattern of an electrical appliance to detect the respective appliance in energy readings from smart meters and smart plugs.

Lastly, we assess the performance of the correlation filters on the real-world energy consumption dataset GREEND, which provides readings from smart meter data as well as appliance-level measurement equipment. As the results approve, the correlation filters show a good performance for appliances with predictable consumption patterns such as refrigerators, dishwashers, or washing machines. Thus, we propose that future work should evaluate the applicability of correlation filters in appliance diagnosis systems."

Christoph Klemenjak presenting at the load classification session

C. Klemenjak and W. Elmenreich. On the Applicability of Correlation Filters for Appliance Detection in Smart Meter Readings. In Proceedings of the 2017 IEEE International Conference on Smart Grid Communications (SmartGridComm), Dresden, Germany, October 2017.


The Correlation Filters were evaluated on real-world energy consumption data provided by the GREEND dataset, which is available at Sourceforge:

A. Monacchi, D. Egarter, W. Elmenreich, S. D'Alessandro, and A. M. Tonello. GREEND: An energy consumption dataset of households in italy and austria. In Proc. IEEE International Conference on Smart Grid Communications (SmartGridComm'14), Venice, Italy, 2014.

Interoperability Between Smart and Legacy Devices in Energy Management Systems

Energy management systems can help to decrease energy consumption by giving user feedback or by directly controlling devices. Smart appliances create a network of devices that can be addressed and controlled via a defined network interface. However, legacy devices will establish a significant portion of a system’s power consumption and, therefore, need to be included into the management system. We propose an open architecture to integrate smart and non-smart devices by using smart plugs and non-intrusive load monitoring methods. Devices are connected either as (i) smart appliances via a fieldbus or wireless network, (ii) legacy devices connected to a smart plug, or (iii) other legacy devices being detected from a time sequence of power consumption values, which are disaggregated into the power draws of different devices. At a service layer, device properties are presented in a unified way including a machine-readable description of their features and properties. The data layer provides an abstract representation of data and functionalities. It connects to the application layer where different applications can access the data. The system supports mechanism for service discovery, service coordination, and service and resource description.

Paper:
D. Egarter, A. Monacchi, T. Khatib, and W. Elmenreich. Integration of legacy appliances into home energy management systems. Journal of Ambient Intelligence and Humanized Computing, 2015.

My talk at Workshop Energieinformatik 45. GI-Jahrestagung "Informatik, Energie und Umwelt":

Slides from Wilfried Elmenreich

European NILM workshop 2015 in London

I'm really happy to here that there will be a Second European Workshop on Non-intrusive Load Monitoring this summer.
The event was announced from Oliver Parson, an organizer of the Workshop, as follows:

I'm really excited to announce that the Second European Workshop on Non-intrusive Load Monitoring will be held on 8th July 2015 at Imperial College London. The workshop is the follow up to last year's NILM @ London workshop, which provided the first European venue which brought together both academics and companies with an interest in energy disaggregation. Updates and registration information can be found at the new website: www.nilm.eu

Some important information:

When: 8th July 2015
Where: Imperial College London, UK
Cost: Free
Objective: To provide a European venue for disaggregation researchers to discuss recent developments in the field and fuel future collaborations

This workshop will have a more technical focus than the first workshop, and will feature a keynote from Mario Bergés in the morning and a technical session of invited talks in the afternoon. Furthermore, we're inviting all attendees to bring a poster on a topic of their choice, which could be a recent piece of work, their company's current direction, or even an invitation for collaboration on a joint project. Last, we're hoping to live stream the event online for anyone who can't be there in person, though this is a little experimental!

For more information please visit www.nilm.eu!

Impressions from IEEE Conference on Smart Grid Communication 2014

The IEEE Conference on Smart Grid Communication 2014 was held from November the 13th to 14th in Venice. Wilfried Elmenreich, Dominik Egarter and Andrea Monacchi from our group participated at this event.

The conference was organized in 5 different symposia:

• Communications and Networks to enable the Smart Grid
• Cyber Security and Privacy
• Architectures, Control and Operation for Smart Grid, Microgrids and Distributed Resources
• Demand Response and Dynamic Pricing
• Data Management and Grid Analytics

Dominik presented his paper Load Hiding of Household's Power Demand, (Dominik Egarter, Christoph Prokop, Wilfried Elmenreich) in the session on "Cyber Security and Privacy".

Andrea gave a talk about his paper GREEND: An Energy Consumption Dataset of Households in Italy and Austria (Andrea Monacchi, Dominik Egarter, Wilfried Elmenreich, Salvatore D’Alessandro, Andrea M. Tonello) in the "Data Management and Grid Analytics" session.

Load Disaggregation Paper accepted in IEEE Transactions on Instrumentation and Measurement

Recently, the paper "PALDi: Online Load Disaggregation via Particle Filtering" got accepted for publication in the journal IEEE Transactions on Instrumentation and Measurement.

Egarter, D., Bhuvana, V. P. & Elmenreich W. (2014). PALDi: Online load disaggregation via particle filtering. IEEE Transactions on Instrumentation and Measurement, pages 467 - 477, 64(2).

Smart metering and fine-grained energy data are one of the major enablers for the future smart grid and improved energy efficiency in smart homes. By using the information provided by smart meter power draw, valuable information can be extracted as disaggregated appliance power draws by non-intrusive load monitoring (NILM). NILM allows to identify appliances according to their power characteristics in the total power consumption of a household, measured by one sensor, the smart meter.

In this paper we present a NILM approach, where the appliance states are estimated by particle filtering (PF). PF is used for non-linear and non-Gaussian disturbed problems and is suitable to estimate the appliance state. On/off appliances, multi-state appliances, or combinations of them are modeled by hidden Markov models (HMM) and their combinations result in a factorial hidden Markov model (FHMM) modeling the household power demand. We evaluate the PF-based NILM approach on synthetic and on real data from a well-known dataset to show that our approach achieves an accuracy of 90% on real household power draws.

Load Hiding of Household's Power Demand

Our publication "Load Hiding of Household's Power Demand" is accepted at the IEEE SmartGridComm 2014. The SmartGridComm 2014 will take place in Venice, a fascinating and marvelous Italian city, on November 3-6, 2014.

Load Hiding of Household's Power Demand, Dominik Egarter, Christoph Prokop, Wilfried Elmenreich

With the development and introduction of smart metering, the energy information for costumers will change from infrequent manual meter readings to fine-grained energy consumption data. On the one hand these fine-grained measurements will lead to an improvement in costumers’ energy habits, but on the other hand the fined-grained data produces in- formation about a household and also households’ inhabitants, which are the basis for many future privacy issues. To ensure household privacy and smart meter information owned by the household inhabitants, load hiding techniques were introduced to obfuscate the load demand visible at the household energy meter. In this work, a state-of-the-art battery- based load hiding (BLH) technique, which uses a controllable battery to disguise the power consumption and a novel load hiding technique called load-based load hiding (LLH) are presented. An LLH system uses an controllable household appliance to obfuscate the household’s power demand. We evaluate and compare both load hiding techniques on real household data and show that both techniques can strengthen household privacy but only LLH can increase appliance level privacy.

An authors' version is available in arXiv: http://arxiv.org/abs/1406.2534

 

Impressions from BuildSys 2013

The 5th ACM Workshop On Embedded Systems For Energy-Efficient Buildings (BuildSys) 2013 were hold from the 13th to 14th of November in Rome, where the two researcher Dominik Egarter and Andrea Monacchi participated in. According to the general BuildSys chair the Workshops' aims can be summarized as follows:

The entrance of La Sapienza University, Rome, where SenSys and BuildSys were held

"BuildSys is a venue for discussing new directions in the monitoring, control, and management of energy consumption in buildings, and the generation of awareness and sustainability for an efficient energy market. BuildSys brings together researchers from a multitude of disciplines with a common goal to develop energy saving strategies that can have a major impact worldwide. BuildSys 2013 follows four successful predecessors held in Berkeley, Zurich, Seattle, and Toronto."

This year the workshop had 7 different season such as:

• Energy Efficiency in Homes,
• Data Analysis,
• Occupancy Detection, Monitoring & Use,
• Sensing for Energy,
• Energy and Water,
• HVAC, Modeling and Control and
• Thermal Comfort Management,

Andrea and Dominik at their posters 

where in total 22 papers were presented. In addition to the general paper presentation, the workshop offers a poster and demo session, where 10 posters and 8 demos were presented. The Smart Grid Group not only visited workshop, Dominik and Andrea were presenting two posters. Dominik showed his current result of Non-Intrusive Load Monitoring (NILM) techniques with the title Appliance State Estimation Based on Particle Filtering. In this paper he showed how it is possible to apply Particle Filtering to the problem of aggregated power loads and how beneficial this approach can be. Moreover, Andreas poster has the title Insert Coin: turning the household into a prepaid billing system and propose an approach for raising energy awareness by combining appliance-level consumption information with prepaid billing so as to turn appliances in pay-as-you-go devices.

D. Egarter, Venkata Pathuri Bhuvana, W. Elmenreich. Appliance State Estimation Based on Particle Filtering, 5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings, Rome, Italy, 2013.

A. Monacchi, W. Elmenreich. Insert Coin: turning the household into a prepaid billing system,5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings (BuildSys'13), Rome, Italy, 2013

Appliance State Estimation based on Particle Filtering

The load disaggregation problem

Dominik Egarter will present his work on Appliance State Estimation based on Particle Filtering at  the 5th ACM Workshop on Embedded Systems for Energy-Efficient Buildings (BuildSys'13) which will take place in Rome, Italy at 13th and 14th October 2013.

Non-Intrusive Load Monitoring is a single-point metering approach to identify and to monitor household appliances according their appliance power characteristics. In this paper, we propose an unsupervised classification approach for appliance state estimation of on/off-appliances modeled by a Hidden Markov Model (HMM). To estimate the states of appliances, we use the sequential Monte Carlo or Particle Filtering (PF) method. The proposed algorithm is tested with MATLAB simulations and is evaluated according to correctly or incorrectly detected on/off events.
On other approaches for solving the NILM problem, see:

Particle filter method

Read more about the approach in:

D. Egarter, Venkata Pathuri Bhuvana, W. Elmenreich. Appliance State Estimation Based on Particle Filtering, 5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings, Rome, Italy, 2013.

On other approaches for solving the NILM problem, see:

D. Egarter and W. Elmenreich. EvoNILM - Evolutionary appliance detection for miscellaneous household appliances. In Proceedings of the Green and Efficient Energy Applications of Genetic and Evolutionary Computation at the 2013 Genetic and Evolutionary Computation Conference (GECCO 2013 GreenGEC). ACM, July 2013.

D. Egarter, A. Sobe, and W. Elmenreich. Evolving non-intrusive load monitoring. In Proceedings of the 15th European Conference on the Applications of Evolutionary and bio-inspired Computation, pages 182–191, Vienna, Austria, April 2013.

Final two Interns@SmartGrid_Group

For the last two months we had further two interns working at our institute. We report a short interview about their work and experiences at our research group.

What's your name?

Julia

What are you studying?

I am studying electrical engineering and industrial electronics at the HTL Mössingerstraße in Klagenfurt and this year I've finished the 4th class.

What are your expectations for the future?

I want to conclude the HTL with good marks, then I would like to study something that combines electrical engineering and economics.

What are you working on?

LabView GUI

I have worked on a LabVIEW measurement program for a photovoltaic system and have also created a consumer box for the lab, where we want to simulate a household. I have installed input modules, which are used for measuring voltage and current, and an output module, which is used for turning the devices on and off. These modules are implemented into my LabView program and further I have created an interface, where the user is able to input data to the program and view power and current consumption. There is also the possibility to read the measured data and control the devices through a web service in the network. I have also installed the hardware elements in the cabinet, created the wiring plan and ordered appliances. The aim of the program is to be used as a control of consumer devices like a refrigerator, water heater and radiator, which are turned on at the time, when the current price is on the lowest peak. Thus this could be an innovative application for saving costs and has to be researched.

Laboratory Installations

What did you like of the project

I liked about my work that I had the chance to contribute my own ideas and expand my programming skills in LabVIEW. It was a very varied project and I liked the mix of manual work and logical thinking.

What's your name?

Johannes

What are you studying?

I'm studying electronics and technical informatics at the HTL Mössingerstraße in Klagenfurt. I have finished my second year.

What are you expectations for the future?

After finishing the HTL, I want to study computer science.

What are you working on?

I was working on Java classes for a Smart Grid Simulator which make it possible to use algorithms from the GridLAB-D simulator. To submit this, I converted all data stored by objects from the Smart Grid Simulator into a file which is then used by GridLAB-D. The output files made by GridLAB-D are read and the values are assigned to the proper object.

OpenEnergyMonitor

GridLAB-D to Smart Grid Simulator 

I was also working on an open source energy monitor made by Openenerymonitor. My task was to enable it to change the refresh rate on the transmitter (emonTX) by using buttons on the receiver (emonGLCD). The difficult thing was to send and receive data at a time using the same transceiver.

Power Profile Generation in Java

My final project was the implementation of a Java program, which generates random power profiles of different appliances out of a given database, place this random profiles randomly in time and finally, tries to detect which appliance was used or not. The technique of appliance detection is called Non-Intrusive Load monitoring and enjoys currently an great interest in research.
What did you like of the project?

I like programming a lot. So I had the opportunity to improve my skills.

Smart Grid @ GECCO 2013

On July 6th - 10th we attended the Genetic and Evolutionary Conference GECCO 2013 in Amsterdam. GECCO is the largest conference in field of genetic and evolutionary computation. The conference was organized by 18 different tracks like Genetic Programming, Genetic Algorithms or Real World Applications, had 35 different tutorials and hosted 13 workshops.

From the Smart Grid group, Dominik Egarter took part at the Green and Efficient Energy Applications of Genetic and Evolutionary Computation Workshop 2013, where he presented his paper EvoNILM - Evolutionary Appliance Detection for Miscellaneous Household Appliances.
In this workshop 5 different papers were presented, which topics were widely spread from power flow optimization, wind power forecasting to wind power siting.

• D. Egarter, W. Elmenreich. EvoNILM - Evolutionary Appliance Detection for Miscellaneous Household Appliances Workshop on Green and Efficient Energy Applications of Genetic and Evolutionary Computation, Amsterdam, July 2013.

Evolutionary Appliance Detection for Miscellaneous Household Appliances

The paper "EvoNILM - Evolutionary Appliance Detection for Miscellaneous Household Appliances" was accepted to the Workshop "Green and Efficient Energy Applications of Genetic and Evolutionary Computation" at Gecco 2013.

To improve the energy awareness of consumers, it is necessary to provide them with information about their energy demand, not just on the household level. Non-intrusive load monitoring (NILM) gives the consumer the opportunity to disaggregate their consumed power on the appliance level. The consumer is provided with information about the energy demand of each individual appliances. In this paper we present an evolutionary optimization algorithm, applicable to NILM purposes. It can be used to detect appliances with a probabilistic power demand model. We show that the detection performance of the evolutionary algorithm can be improved if the single population approach of the evolutionary algorithm is replaced by a parallel population approach with individual exchange and by the introduction of application-oriented pre-processing and mutation methods. The proposed algorithm is tested with Matlab simulations and is evaluated according to the fitness reached and detection probability of the algorithm.

This paper is an improvement and follow up paper of the previous work "Evolving Non-Intrusive Load Monitoring".

• D. Egarter, W. Elmenreich. EvoNILM - Evolutionary Appliance Detection for Miscellaneous Household Appliances Workshop on Green and Efficient Energy Applications of Genetic and Evolutionary Computation, Amsterdam, July 2013.

Evolving Non-Intrusive Load Monitoring

Our paper Evolving Non-Intrusive Load Monitoring by Dominik Egarter, Anita Sobe and Wilfried Elmenreich has been accepted for the conference track EvoEnergy (Evolutionary Algorithms in Energy Applications) of the EvoApplication (16th European Conference on the Applications of Evolutionary Computation) 2013, taking place in Vienna form 3rd to 5th of April.

Basic principle of the ON/OFF time genome appliance
detection. Given is the total power consumption over
time. The goal is to deduce the on/off times of devices
(colored blocks) that add up to the measured power profile.

Non-intrusive load monitoring (NILM) identifies used appliances in a total power load according to their individual load characteristics. In this paper we propose an evolutionary optimization algorithm to identify appliances, which are modeled as on/off appliances. We evaluate our proposed evolutionary optimization by simulation with Matlab, where we use a random total load and randomly generated power profiles to make a statement of the applicability of the evolutionary algorithm as optimization technique for NILM. Our results shows that the evolutionary approach is feasible to be used in NILM systems and can reach satisfying detection probabilities.



Dominik Egarter, Anita Sobe, Wilfried Elmenreich,   Evolving Non-Intrusive Load Monitoring,   EvoApplication 2013,   16th European Conference on the Applications of Evolutionary Computation, April, 2013