Thursday, March 14, 2019

Open thesis topics

Analysis of dataset for energy forecasting

Keywords: RES, energy forecasting, training dataset, testing dataset,  prediction error

Description and objectives:

Integration of Renewable energy resources (RES) in today's power grid depends highly on the quality of energy forecasting outcome. However, the variability of RES poses several challenges to this integration. The idea of the proposed thesis is to assess the influence of different dataset composition which includes varying training, validation and testing dataset on the prediction error and to come up with the most effective training method. The performance of the training method will be validated over publicly available dataset using available performance evaluation metrics.

Main tasks:

  • The first task is to look for dataset having different characteristics in order to achieve better comparison and assessment of most effective approach when applied to different datasets.
  • The next task is to train neural network using machine learning algorithms for different dataset composition. 

Qualifications: Programming skills in Python

Contact Details:  Ekanki SHARMA : 

Impact of data pre-processing techniques on forecasting accuracy

Keywords: RES, data pre-processing, energy forecasting, feature selection, outlier rejection, forecasting accuracy

Description and objectives:

To integrate renewable energy sources (RES) in power grid, forecasting the photovoltaic (PV) yield is very important. Several techniques have been implemented in the literature which includes naive (time-series, statistical) methods to soft computing techniques (Artificial neural network, support vector machine, grey prediction) to improve the accuracy of the forecasting model. The idea of the thesis is to evaluate the impact of applying feature selection and outlier rejection techniques on forecasting accuracy. 

Main tasks:

  • The first task is to search for available pre-processing techniques.
  • The next step is to compare the forecast accuracy with and without applying data pre-processing techniques.
Qualifications: Programming skills in python

Contact Details:  Ekanki SHARMA : 

Thursday, March 1, 2018

Meetup on Future Research Challenges in Energy Informatics

18 - 19 April 2018 @ University of Klagenfurt, Austria


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.


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.

Friday, December 1, 2017

European Workshop on Non-Intrusive Load Monitoring (NILM)

From 6th to 7th of November, the 4th European Workshop on Non-Intrusive Load Monitoring (NILM) was held in London, United Kingdom. This event brought together researchers and professionals to present and discuss latest developments in the area of NILM as well as its applications. 

The sessions comprised topics such as commercial & industrial NILM, innovative algorithms, deep learning approaches, and evaluation. Also, several vendors such as VoltawareQualisteo or Verv introduced their latest products.

In the poster & demo session, Christoph presented a poster on "Appliance Detection in Power Meter Readings". The poster illustrates how correlation can be utilised to detect electrical appliances in power readings. Especially for hardware with limited computational resources this approach shows promising results. For more information about the presented work refer to our paper on correlation filters for appliance detection.

Picture: Christoph Klemenjak (on the left) presenting his poster on appliance detection

A selection of the workshop's talks is available in form of a Youtube playlist.

The next major NILM event will be the International Workshop on Non-Intrusive Load Monitoring, which will take place in late February / early March 2018 in Austin, Texas.

Wednesday, November 22, 2017

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.


"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.

Friday, November 10, 2017

Should I drive with my car to turn off forgotten lights?

I forgot to turn off the lights in my office today. True story. And it's Friday, so unless the cleaning personel turns them off, the lights there will burn unnecessarily for 60 hours until Monday morning. This is a bit embarassing when you are doing research[1] and teaching[2] in sustainability and energy management.
So the question is, should I go there immediately and turn them off? Normally I would use my bike, but it's dark and cold outside so I consider using the car.

Let's crunch the numbers first. My car consumes about 5,5l Diesel per 100 km, so using it for a single person trip is far from contributing to a sustainable future. But the trip will save the energy that would be consumed by the office lights. The trip there and back is 11km, so this would use 0.6 l of Diesel. 1 liter of diesel contains chemical energy worth 9.85 kWh, weighs 835 g, and contains 86% carbon. This 720 g carbon would be mostly burned to CO2, resulting in 2640 g of CO2 [3].

Burning 0,6 l Diesel would thus generate about 1,5 kg of CO2 and waste 5,9 kWh of energy.

What if I let the lights burn? Unfortunately the office lights are not LED-based, but they are fluorescent tubes. I would guess all together the lighting has a power consumption of 100 W. Letting them burn for 60 h would thus waste 6 kWh, basically the same value that we calculated for the used fuel.

About 3/4 of Austria's "Strom-Mix" come from hydropower, wind, waste and solar sources, the remaining part from fossil fuels like gas, oil and coal. I'm lucky that Austria has no nuclear power, first because it is a dangerous technology and second because the effective CO2 emissions of nuclear power are hard to estimate :-).

Coal is the worst source, it comes with 882 g CO2 per kWh[4]. All together Austria's electric energy comes with emissions of 181 g/kWh[5]. So the 6 kWh of electrical energy from the "let's the lights burn" scenario are a bit more than 1kg - less than the scenario where I drive with the car to turn the lights off.

So I should feel bad about the environment, but at least I have my lazyness supported. I might go there by bike tomorrow :-)

Saturday, July 29, 2017

Smart Microgrids and Renewable Energies in Smart Grids group AAU

A smart grid is a modernized electric power grid that uses information and communication technologies (ICT) to improve the reliability, security, sustainability, and efficiency of an electric power system. One key characteristic of a smart grid is to provide two-way information exchange via communication technology. This allows building applications that link producers and consumers for a better planning of resources, new comfort features, and for integrating renewable energy sources into the grid.
To realize power grids with a high number of renewable energy resources, a number of research problems need to be solved. Out of all, balancing production and consumption in such a power grid is a major issue, as renewable energy resources are intermittent in nature- their output heavily depends on weather conditions.With increased focus on this integration of renewable energy resources, a framework of smart microgrids is a viable approach to grid modernization. A smart microgrid can be described as a decentralized grouping of electricity generation, energy storage, and loads. Usually, the generation and loads in a microgrid are connected at low-voltage (LV) level and medium-voltage (MV) level. 

Professor Elmenreich and his team work on different areas of Smart Microgrids which include: 
  • Simulation of smart microgrids
  • Prediction models for photovoltaic systems

Due to the unpredictable nature of photovoltaics and wind power and their dependence on meteorological conditions, it is important to model and simulate these sources in power system to study their impact on power flow and the quality of the power system.

Renewable Alternative Power System Simulator- RAPSim

Professor Elmenreich and his team developed a Renewable Alternative Power System Simulator- RAPSim. "RAPSim is a free and open-source software, microgrid simulation framework to better understand the power flowing behavior in smart microgrids with renewable sources and load demands. It is able to simulate grid-connected or standalone microgrids with solar, wind or other renewable energy sources," explains researcher Manfred Rabl-Pöchacker.

In order to analyze the impact of added renewable energy sources on the microgrid variables, power flow analysis is important. The proposed software allows to define the power generated or consumed by each source in the microgrid and then provides a power flow analysis. RAPSim is designed for application in science and classroom with a simple to use graphical interface. It is an easily extendable framework that supports users in implementing their own models, in particular, grid objects such as new types of producers or consumers and algorithms for grid control.

RAPSim Video

Prediction Models for Photovoltaics 

Photovoltaic (PV) systems have received a lot of attention due to their ecological property of efficiently converting the usable solar power into electricity. “Forecasting the PV output power of smart microgrid is essential for an efficient use of an electricity grid.,“ states Ekanki Sharma. She is currently working on different aspects of solar power forecasting techniques.

The team lead by Elmenreich has investigated how computational methods and principles can assist in planning smart microgrids. “In a recent case study, we trained a neural network with sensor data as well as with energy production data of renewable energy plants. The results indicate that neural networks are able to forecast the production of photovoltaic and wind power plants,“ reports Professor Elmenreich.

The group also contributed towards machine learning applications, in particular for modeling solar radiation. They presented the possibility of using an Artificial Neural Network (ANN) to grasp the correlation of input parameters and predict global and diffuse solar radiation. Experiments revealed that applying ANN can achieve higher accuracy than existing models.

Selected Publications 

M. Pöchacker, T. Khatib, and W. Elmenreich. The microgrid simulation tool RAPSim: Description and case study. In Proc. IEEE Innovative Smart Grid Technologies Asia, 2014.

T. Khatib, and W. Elmenreich. Novel simplified hourly energy flow models for photovoltaic power systems. Energy Conversion and Management, 2014.

Modeling of Photovoltaic Systems Using MATLAB: Simplified Green Codes

When investigating the field of photovoltaic systems as a student, one faces the situation that, despite a high number of publications on the topic, it is hard to find out how to start with your own experiments and simulations. To address this issue, the book Modeling of Photovoltaic Systems Using MATLAB: Simplified Green Codes (Wiley, 2016) describes models of photovoltaic systems interleaved with short MATLAB programs showing how to set up a simulation based on the content of the current chapter. The chapters of the book cover different systems including photovoltaic energy sources, storage, and power electronic devices.
The book resulted from the collaboration of Tamer Khatib and Wilfried Elmenreich at the NES Institute in the years 2013 to 2015.

Book Preview

Smart Microgrid Lab

With the roll-out of smart grid technology, the importance of metering, communication, and distribution has risen rapidly. To give students a hands-on experience on research activities going on in the domain of smart grids, the Smart Microgrid Lab was established. The lab is part of the Lakeside Labs GmbH and is operated in cooperation with the Smart Grids Group at Alpen-Adria-Universität Klagenfurt.

Figure: Lucas-Nülle Advanced Photovoltaics Trainer 

Experimental setups include the combination of simulation scenarios with actual hardware in the loop. This way, students, and researchers can perform experiments and test modeling an entire power grid from power generation to power consumption. A special focus is given to the integration of photovoltaics so that students can design their own microgrid. 

The lab is connected to a Photovoltaic (PV) plant installed on the rooftop of the building with a capacity of 4.8kWp, which compares to a PV system for typical single-family houses. Several test loads with common household appliances are available in the lab to perform experiments in this context. 

Usually, this system is connected to the grid, but in the case of a power outage, the lab can sustain itself in an off-grid (island) mode. “If everything goes dark, students can still continue learning in our lab,” Elmenreich adds with a wink. 

Friday, July 21, 2017

Energy Informatics in Klagenfurt at a Glance

In the present day, the electric power grid faces an evolutionary step towards the smart grid. The smart grid is defined as the enhancement of the electric power grid with information and communication technology. This sort of digitisation will enable a bidirectional flow of energy and information within the power grid and provide several novel applications and allow to unlock the full potential of renewable energy technologies. To cope with the challenge of digitisation in power grids, key elements of future energy systems have to be explored and furthermore, computational methods have to be developed and refined.

The Smart Grids group, located at the University of Klagenfurt, contributes to this challenge by investigating how power meter readings can be analysed to discover solutions to sustainably increase the energy efficiency of energy systems. 

"We carried out a measurement campaign in eight selected households to track power consumption of individual electrical appliances for over one year. The main outcome was the GREEND dataset, which was analysed to gain further insights into energy consumption behaviour", says Andrea Monacchi, the coordinator of the campaign. 

Figure: Overview of the households and appliances

On basis of such data, policies can be formulated to improve energy efficiency by shedding of standby losses, postponement to off-peak periods, replacement of inefficient appliances, and operation curtailment. The networked power meter (smart meter) represents the key element in the transition towards the smart grid since such measurement equipment provides feedback to the users, other appliances, and the electric utility. 

"The smart meter is a vital tool for researchers to record the energy consumption of households and industrial buildings. On basis of the collected data, computational methods and effectivity of efficiency measures are evaluated", states Christoph Klemenjak. "We developed an open-hardware smart metering board called YoMo, which is an extension unit for the Arduino platform. The YoMo is designed to monitor current flow and voltage level, as well as active, reactive, and apparent power at the feed point of households."

Figure: The YoMo smart metering board

In general, smart meters serve several purposes such as billing of consumed energy, providing immediate feedback to the users, or switching loads. Smart metering and fine-grained energy data are one of the major enablers for the future smart grid and an improved energy efficiency in smart homes. On the one hand these fine-grained measurements will lead to improved energy consumption habits, on the other hand the fine-grained data produces many questions with respect to privacy issues. To ensure both, household privacy and smart meter information, load hiding techniques were introduced to obfuscate the load demand visible at the household energy meter. 

"The load hiding technique we developed works with devices that are already in the system. By controlling these household appliances in a certain way, your power profile becomes scrambled", Professor Elmenreich explains. 

Selected Publications 

C. Klemenjak, D. Egarter, and W. Elmenreich. YOMO-The Arduino based smart metering board. Computer Science - Research and Development (Springer), 2016. 

M. Pöchacker, D. Egarter, and W. Elmenreich. Proficiency of power values for load disaggregation. IEEE Transactions on Instrumentation and Measurement, 2016. 

D. Egarter, V. P. Bhuvana, and W. Elmenreich. PALDi: Online Load Disaggregation based on Particle Filtering. IEEE Transactions on Instrumentations and Measurement, 2015. 

D. Egarter, C. Prokop, and W. Elmenreich. Load Hiding of Household's Power Demand. IEEE International Conference on Smart Grid Communications, 2014. 

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, 2014. 

The Energy Informatics Lab

Data will be one of the most important resources in the future. Among other timely research questions, activities in energy informatics explore how data provided by advanced metering infrastructure (AMI) can be utilised in the most adequate and efficient way. In order to record energy data, the Energy Informatics Lab provides a wide range of measurement instruments such as smart meters, oscilloscopes, and self-designed power meters. Furthermore, modern computer infrastructure allows to evaluate novel computational methods for prediction and user feedback for future energy systems.

Figure: A small selection of the present measurement equipment

The Energy Informatics lab also integrates a well-equipped soldering work station and a 3D printer, which allows the researchers to craft prototypes and custom enclosures for all kinds of measurement equipment. “Previously our students often had problems apply the theoretic concepts they learned. The Energy Informatics Lab offers the possibility to try out things in practice, which gives them a different perspective,” states Professor Wilfried Elmenreich.

Figure: The present 3D printer allows rapid prototyping

More detailed information can be found in our research blogs Energy InformaticsThe Smart Grid, and on the institute's web page Networked and Embedded Systems.