Program

Wednesday, October 10
From 18:00 Welcome Reception
 Thursday, October 11
9:00 – 9:30 Welcome Address
9:30 – 10:30 Keynote
10:30 – 11:00 Coffee Break
11:00 – 12:30 Session 1: Home Energy Management
12:30 – 13:00 Posters & Demos: 1-Minute Madness
13:00 – 14:30 Lunch
14:30 – 16:00 Session 2: Network Operations
16:00 – 16:30 Coffee Break
14:30 – 16:30 Tools & Methods Workshop
16:30 – 17:30 Session 3: Posters & Demos
17:30 – 17:45 Information about Dinner Logistics
From 19:00 Social dinner and announcement of the winner of the Best Paper Award
Friday, October 12
9:00 – 9:30 Welcome Session and announcement of Energieinformatik 2019
9:30 – 10:30 Keynote
10:30 – 11:00 Coffee Break
11:00 – 12:30 Session 4: Optimization & Interoperability
12:30 – 13:30 Lunch
13:30 – 15:00 Session 5: Use Case-based Analysis
15:00 – 15:30 Wrap-up and farewell
From 16:00 SINTEG enera Open Space visit

 

Keynotes

Agility, Stability and Complexity: The Challenges of Security in Critical Energy Infrastructures

ABSTRACT: Maneuvering highly dynamic, unpredictable, complex and contradictory environments requires a qualitatively new flexibility from energy companies. Customers, suppliers and employees continue to expect maximum stability, but this can only be offered – paradoxically – if the organization is able to react agile “just in time” to new requirements. In order to be able to offer themselves and others stability, employees, management, structures and processes must become flexible. This keynote highlights this field of tension and presents good examples and failures.


Fotograph: Luca Melette

Linus Neumann

Linus Neumann is a computer hacker and psychologist. He is a spokesperson of the Chaos Computer Club, Europe largest hacker collective. Furthermore, Linus Neumann publishes the podcast Logbuch:Netzpolitik together with Tim Pritlove.

IT-Security for Future Energy Systems: A Lost Cause?

ABSTRACT: Making energy systems more intelligent brings undisputed advantages and is seen as an important enabler as we turn from fossil to renewable energy sources. However, there are severe concerns regarding the security of these complex systems. Introducing information and communication technology not only paves the way for new and important functionality, but also opens the door for new vectors of cyber attacks that can potentially harm this critical infrastructure. Critics see the cyber defense of future energy systems as a lost cause, as the systems are perceived as too complex to be secured properly. In this talk, the question is addressed, if systems as complex as future energy systems can be secured from cyber attacks at all –- and if so, what are the most promising approaches. Current threats and trends in cyber security for energy systems are reviewed. The state of the art in counter-measures, including new approaches, such as blockchain technology, and topics such as the interplay of IT-security measures with end-user privacy are discussed. Finally, an outlook on research challenges is given.


Fotograph: Hannelore Kirchner

Prof. (FH) Dr. Dominik Engel

Dominik Engel is a professor at the Salzburg University of Applied Sciences in Austria, where he heads the Center for Secure Energy Informatics. He holds a PhD degree in Computer Science from the University of Salzburg. Prior to joining Salzburg University of Applied Sciences, Dominik Engel was a researcher at the Universities of Bremen and Salzburg and product manager at Sony DADC, where he was responsible for video content security. His current research interests include smart grid privacy and security and technological methods for enhancing end-user trust. Dominik Engel has authored and co-authored a number of publications related to security and privacy in smart grids and is a member in various EU and national standardization committees in this area.

Session 1: Home Energy Management

Session chair: Prof. Dr. Anke Weidlich (University of Freiburg, DE)

Calculating Retail Prices from Demand Response Target Schedules to Operate Domestic Electric Water Heaters
Tobias Lübkert, Marcus Venzke and Volker Turau (all Hamburg University of Technology, DE)
The paper proposes a demand response scheme controlling many domestic electric water heaters (DEWHs) with a price function to consume electric power according to a target schedule. It discusses at length the design of an algorithm to calculate the price function from a target schedule. The price function is used by the control of each DEWH to automatically and optimally minimize its local heating costs. It is demonstrated that the resulting total power consumption approximates the target schedule. The algorithm was successfully validated by simulation with a realistic set of 50 DEWHs assuming perfect knowledge of parameters and water consumption. It is shown that the algorithm is also applicable to clusters of large numbers of DEWHs with statistical knowledge only. However, this leads to a slightly higher deviation from the target schedule.

Estimating the Savings Potential of Occupancy-based Heating Strategies
Vincent Becker, Wilhelm Kleiminger, Vlad C Coroama and Friedemann Mattern (all ETH Zurich, CH)
Because space heating causes a large fraction of energy consumed in households, occupancy-based heating systems have become more and more popular in recent years. However, there is still no practical method to estimate the potential energy savings before installing such a system. While substantial work has been done on occupancy detection, previous work does not address a combination with heating simulation in order to provide an easily applicable method to estimate this savings potential. In this paper we present such a combination of an occupancy detection algorithm based on smart electricity meter data and a building heating simulation, which only requires publicly available weather data and some relevant building characteristics. We apply our method to a dataset containing such data for several thousand households and show that when taking occupancy into account, a household can save over 9% heating energy on average, while certain groups, such as employed single-person households, can even save 14% on average. Using our approach, households with high potential for energy savings can be quickly identified and their inhabitants could be more easily convinced to adopt an occupancy-based heating strategy.

Modeling Flexibility Using Artificial Neural Networks
Kevin Förderer (FZI, DE), Mischa Ahrens, Kaibin Bao, Ingo Mauser (all KIT, DE) and Hartmut Schmeck (FZI, DE, and KIT, DE)
The exibility of distributed energy resources (DERs) can be modeled in various ways. Each model that can be used for creating feasible load profiles of a DER represents a potential model for the flexibility of that particular DER. Based on previous work, this paper presents generalized patterns for exploiting such models. Subsequently, the idea of using artificial neural networks in such patterns is evaluated. We studied different types and topologies of ANNs for the presented realization patterns and multiple device configurations, achieving a remarkably precise representation of the given devices in most of the cases. Overall, there was no single best ANN topology. Instead, a suitable individual topology had to be found for every pattern and device configuration. In addition to the best performing ANNs for each pattern and configuration that is presented in this paper all data from our experiments is published online. The paper is concluded with an evaluation of a classification based pattern using data of a real combined heat and power plant in a smart building.

Household CO2-Efficient Energy Management
Laura Fiorini (University of Groningen, NL) and Marco Aiello (University of Groningen, NL, and University of Stuttgart, DE)
Residential and commercial buildings are responsible for one third of the total carbon dioxide (CO2) emissions in the European Union, which are the main cause of global warming. Although the thermal load has long been considered the primary reason of domestic energy consumptions, the increasing demand for electricity has a non-negligible environmental impact, given that about 40% of electricity is generated by burning fossil fuels. Moreover, the amount of CO2 emitted to produce one kWh can greatly vary in time, depending on the sources used to generate it. For instance, the German electricity emissions intensity factor varied in 2017 between 113 and 533 gCO2eq/kWh. This paper proposes a novel CO2-efficient energy management approach to schedule household appliances while minimizing carbon dioxide emissions, given the possibility to change energy carriers (i.e., natural gas and electricity) and to shift loads in time. Several common loads are considered, and their operation is scheduled according to the emission factor of the German power grid. The results show that switching energy carriers can successfully enable up to 40% emissions reductions while indicating that shifting loads in time has little impact.

Session 2: Network Operations

Session chair: Prof. Dr. Hartmut Schmeck (KIT, DE)

Enhancing Power Quality in Electrical Distribution Systems Using a Smart Charging Architecture
Ammar Alyousef, Dominik Danner (both University of Passau, DE), Friederich Kupzog (AIT, AT) and Hermann de Meer (University of Passau, DE)
The electrification of the mobility sector comes with multiple challenges such as the lack of information on when, where, how long and how fast charging processes of electric vehicles will take place. In order to keep up with increasing power demand of charging processes, besides better predictions also the active control of charging processes will be necessary to minimize infrastructure costs. This work deals with a real-time mechanism for supporting the Power Quality (PQ) in electric distribution grids in terms of congestion and voltage management. In the paper, we propose a distributed smart charging approach that considers real-time conditions of the distribution grid provided by an event-driven architecture that collects data from different points in the grid. Our approach adopts the traffic light model, which allows smooth changing of the charging power to avoid drastic changes of the grid state. In order to be ready for real-world application, the algorithm is validated by a series of experiments on two setups: Pure software (co-)simulation and Power Hardware In the Loop (PHIL) where physical charging stations and electric cars are controlled in a laboratory setup.

Systematic Dynamic Assessment for Resilient Operation of Distribution Networks
Felipe Castro, Jorge Velásquez, Davood Babazadeh and Sebastian Lehnhoff (all OFFIS, DE)
The integration of decentralized energy resources is changing the ways distribution grids are being operated. Moreover, a major change is the implementation of active control strategies to adapt to the energy transition, such as voltage and reactive power in-feed control. Correspondingly, the dynamic nature of these controllers, the potential conflict of control objectives and uncertainty of these sources require frequent assessment of the system’s dynamic behaviour. For this reason, this paper defines high level guidelines that allow simplifying Dynamic Security Assessment (DSA) in distribution networks, through the implementation of a methodology to determine the interactions of the network with its controllers. With this in mind, the focus of these guidelines is kept in reducing the computational effort, by reducing the number of dynamic simulations, as well as the amount of active controllers necessary for DSA. These guidelines are verified in an artificial 242-busbar distribution network, resulting in a reduction of 21,4% in number of dynamic simulations required for DSA.

Ensuring Usability of Future Smart Energy Control Room Systems
Tilo Mentler, Tim Rasim, Marcel Müßiggang and Michael Herczeg (all University of Luebeck, DE)
Energy providers face several technical and societal challenges with respect to renewable energies and smart energy systems. As central management units of energy supply systems, control rooms and their operators are especially affected by those changes. While reliability and safety of software systems for managing electric power grids is of utmost importance, their usability has to be ensured as well in order to allow for safe and efficient operations. Previous work has failed to address issues of work reengineering and user interface design for future smart energy control rooms due to insufficient collaboration of human-computer interaction researchers, energy informatics researchers and energy sector stakeholders. This paper describes challenges and approaches for ensuring usability of future smart energy control room systems. It is based on a human-centered design process within an interdisciplinary research project bringing together the aforementioned groups of experts. Results were derived from systematic literature review, workshops and surveys with control room operators as well as contextual inquiries in three control rooms. They concern both the process of realising software systems for managing electric power grids and applications characteristics with respect to user interface design. It is concluded that open and modular software systems require consistent user interfaces based on a style guide. Furthermore, software and usability engineering processes of energy control systems have to be aligned in order to ensure usability, safety and security.

Distributed ledger technology for fully automated congestion management
Astrid Nieße (LUH, DE), Norman Ihle, Stephan Balduin (both OFFIS, DE), Matthias Postina (EWE, DE), Martin Tröschel and Sebastian Lehnhoff (both OFFIS, DE)
Congestion management in distribution grids is an important task for distribution grid operators, both from a financial and a technological perspective. Whereas large generation units and large controllable loads might in general be controllable in a manual way, this is no option for small distributed generators and loads. With flexibility control in multiple owner scenarios, documentation, transparency and automation are of crucial importance. In this work, we present a fully automated congestion management approach based on a combination of distributed ledger technology and distributed algorithms in an agent-based architectural approach. We present a case study focused on the visualization of the concept and discuss the advantages and possible challenges for this approach. Whereas distributed ledger technology has been introduced for peer-to-peer energy trading within the last years, no similar approach has been presented yet for stable distribution grid management.

Tools & Methods Workshop

Organizing committee
Prof. Dr. Sebastian Lehnhoff (OFFIS, DE)
Prof. Dr.-Ing. Astrid Nieße (LUH, DE)

For the last nine years, the PhD workshop Energy Informatics has been a successful forum for young researchers to present their projects, the methods/solutions developed, the tools used in the context of future energy systems. Very different aspects of Energy Informatics (EI) have been addressed over these years, from specific applications such as grid operation, EV charging management or VPP scheduling to systematic topics and architectures like prototyping platforms and standardization processes. Whereas most of the former workshop participants have finished their thesis and are now pursuing interesting careers in EI, some of the tools and methods developed during that work have persisted and since then been further enhanced and established in the respective research groups’ set of tools and methods, i.e. the state of the art in Smart Grid research.

In this workshop on “Tools and methods in Smart Grid research” we aim at presenting some of these in the form of “best practice“ examples. For this reason, we invited EI research groups to present their „bread-and-butter” tools and methods used for Smart Grid research.  The workshop will include the following talks:

G-DPS Framework: Game-theoretical Decision-making for Physical Surveillance Games
Mohamed Amine Abid, Ali Alshawish and Hermann de Meer (all University of Passau, DE)
Protecting critical infrastructure is becoming a major concern for critical infrastructures such as utility networks. Surveillance technologies represents a standard practice for protecting such infrastructures. However, depending on how they are configured, surveillance systems are prone to technical or organizational failures resulting in imperfect performance. Hence, the core problem is to find an optimal configuration of the surveillance technology at hand to minimize such risk.
In this work, we present two tools that were developed in Hyrim project. They are used in our proposed decision-making framework, which assesses possible choices and alternatives towards finding an optimal surveillance configuration and hence minimizing addressed risks. The first tool, the Hyrim Tool, is a game-theoretic model for optimizing physical surveillance systems and minimizing the potential damage caused by an intruder with respect to the imperfect detection rates of the surveillance technology. The second tool, the surveillance simulator, serves to assess the effectiveness of the different surveillance (defense) strategies. It simulates realistic physical intrusion scenarios, and assess the effectiveness of the deployed defense strategies.

Blockchain-Based Management of Switchable Loads
Manuel Utz, Simon Albrecht and Jens Strüker (all Fresenius Universiy of Applied Sciences, DE)
The blockchain technology has been subject to inflated expectations and super ficial research throughout the last two years. Several publications have identifi ed research gaps regarding the real-world deployment of blockchain ecosystems in field tests. We address these gaps by our approach to realize a blockchain demonstrator with tangible components and devices connected to a blockchain platform.
In this paper we document the creation of such a demonstrator. In particular, we deploy blockchain-connected power outlets with three different functionality levels in a smart home environment. 1) The gray outlet allows for precise billing and proof-of-origin of consumed electricity. 2) The green outlet enables dynamic load switching based on dynamic rates and based on the current index of locally available renewable energy. 3) The blue outlet acts as a wallet and switches devices based on blockchain transactions. Data from each outlet is distributed using NFC chips and Rasberry PIs on which flow-based programming tools for IoT applications are installed. A multitude of Python scripts ensures the tracking and categorization of transactions between the three power outlets and household appliances.
From our work we can derive practical implications for the wide-scaled implementation of blockchain-connected devices in residential homes in order to safeguard the incentivization of grid-stabilizing and ecological behavior.

Rapid Prototyping with i7-AnyEnergy and Detailed Co-Simulation with SGsim
Reinhard German, Peter Bazan (both FAU, DE) and Abdalkarim Awad (Birzeit University, PS)
After an overview of simulation approaches we will present two simulation frameworks: i7-AnyEnergy is a library for the commercial simulation tool AnyLogic and allows for rapid prototyping of simulation models of connected intelligent energy systems. It offers predefined model components from which complex system models can be composed easily. The basic components are models for electric and thermal demands (e.g., gas heating, CHP plant), renewable energies (e.g., solar or wind power), energy storages (batteries, chemical storages), intelligent controllers as well as weather models. Simulation models of single houses, virtual batteries as well as a cement plant will be shown. SGsim is a co-simulation framework consisting of the electricity network simulator OpenDSS (for e.g. transmission lines, transformers, generators, loads) and the packet-level computer network simulator OMNeT++ (the INET frameworks allows for the inclusion of detailed communication protocols) . The interface is based on the Component Object Model (COM) or on Object Linking and Embedding (OLE) such that components like supplies, loads, controllers and more can have representations on both sides. Links to external optimizers and phasor data concentrators are possible as well. As a modeling example, volt/var optimization via wireless communication will be shown.

Session 3: Posters & Demos

Session chair: Michael Brand (University of Oldenburg, DE)

(1) OpenDISCO – Open Simulation Framework for Distributed Smart Grid Control
Marius Stübs and Kevin Köster (both University of Hamburg, DE)

(2) A Platform for Testing the Performance of Metaheuristics Solving the Energy Resource Management Problem in Smart Grids
Fernando Lezama, Joao Soares and Zita Vale (all Polytechnic of Porto, PT)

(3) Prediction of domestic appliances usage based on electrical consumption
Patrick Huber, Mario Gerber, Andreas Rumsch and Andrew Paice (all Lucerne University of Applied Sciences and Arts, CH)

(4) Use Case methodology: a progress report
Marie Clausen (OFFIS, DE), Rolf Apel (Siemens, DE), Marc Dorchain (Software AG, DE), Matthias Postina (EWE, DE) and Mathias Uslar (OFFIS, DE)

(5) Flexible and Reconfigurable Data Sharing for Smart Grid Functions
Carsten Krüger, Jorge Velasquez, Davood Babazadeh and Sebastian Lehnhoff (all OFFIS, DE)

(6) Design and Implementation of a Blockchain Multi-Energy System
Qianchen Yu (ETH Zurich, CH), Arne Meeuw and Felix Wortmann (both University of St. Gallen, CH)

(7) Lessons Learned from CPES Co-Simulation with Distributed, Heterogeneous Systems
Cornelius Steinbrink (OFFIS, DE), Christian Köhler (Venios, DE), Marius Siemonsmeier (RWTH Aachen, DE) and Thorsten van Ellen (BTC, DE)

(8) Time Series Analysis with Apache Spark and its Applications to Energy Informatics
Cornelia Krome and Volker Sander (both FH Aachen, DE)

(9) Faster switching of energy suppliers – A blockchain-based approach
Michael Hinterstocker (Forschungsgesellschaft für Energiewirtschaft, DE), Florian Haberkorn (Forschungsgesellschaft für Energiewirtschaft, DE, and Ulm University, DE), Andreas Zeiselmair (Forschungsstelle für Energiewirtschaft, DE) and Serafin von Roon (Forschungsgesellschaft für Energiewirtschaft, DE)

(10) Realistic modeling of a combined heat and power plant in the context of mixed integer linear programming
Thomas Weber, Nina Strobel, Thomas Kohne, Jakob Wolber and Eberhard Abele (all TU Darmstadt, DE)

(11) Enhancing Neural Non-Intrusive Load Monitoring with Generative Neural Models
Kaibin Bao, Kanan Ibrahimov, Martin Wagner and Hartmut Schmeck (all KIT, DE)

(12) Structured workflow achieving interoperable Smart Energy systems
Marion Gottschalk (OFFIS, DE), Gerald Franzl (AICO-Software, AT), Matthias Frohner, Richard Pasteka (both University of Applied Science Technikum Wien, AT) and Mathias Uslar (OFFIS, DE)

(13) A Context-Based Building Security Alarm through Power And Sensors Analysis
Francisco Silva, Gabriel Santos, Isabel Praca and Zita Vale (all Polytechnic of Porto, PT)

(14) Analyzing controller conflicts in multimodal smart grids – framework design
Astrid Nieße, Arash Shahbakhsh (both LUH, DE)

(15) Energy forecasting based on predictive data mining techniques in Smart Energy Grids
Ekanki Sharma (University of Klagenfurt, AT)

(16) On performance evaluation and machine learning approaches in non-intrusive load monitoring
Christoph Klemenjak (University of Klagenfurt, AT)

(17) Blockchain-based orchestration of distributed assets in electrical power systems
Jonas Schlund (FAU, DE)

(18) Methodology for controller interaction assessment in distribution networks with a high share of renewable energy
Jorge Velasquez (OFFIS, DE)

(19) Agent-based dynamic optimization of local controller configurations in converter dominated electricity grids using decoder functions
Johannes Gerster (OFFIS, DE)

(20) Robustness Modeling and Assessment of Interdependent Smart Grids
Ferdinand von Tüllenburg (Salzburg Research, AT)

(21) Surrogate models for composed simulation models in energy systems
Stephan Balduin (OFFIS, DE)

(22) Towards optimized exchange topologies in smart distribution grids
Stefanie Holly (OFFIS, DE)

Session 4: Optimization & Interoperability

Session chair: Prof. Dr.-Ing. Astrid Nieße (LUH, DE)

Scaling: Managing a large number of distributed battery energy storage systems
Hubert Abgottspon, René Schumann (both HES-SO Valais/Wallis, CH), Lucien Epiney and Karl Werlen (both Misurio AG, CH)
This paper analyzes the management of a large number of distributed battery energy storage systems (BESSs) by a energy utility in order to provide some market services. A heuristic algorithm based on two parts is proposed for this task. The first part, the aggregation, combines the abilities and behavior of the fleet of BESS into a virtual power plant (VPP) by a concise but flexible model. This VPP can be used by the utility as they are used to with traditional power plants. The second part, the disaggregation, distributes VPP control schedules back to the individual BESS by a greedy first-fit decreasing heuristic. The management of a fleet of BESS can also be modeled as a mathematical linear optimization program. The proposed heuristic is compared to and evaluated against this global optimization regarding computational performance and quality of results. It is shown, that the heuristic provides a remarkable speedup when applied to larger number of units. With it, it is possible to handle a group of at least 100’000 individual BESS. Further, the quality of the results are shown. First, the solution of the heuristic is compared to the optimal one of the mathematical program. Second, the methods are both applied and compared in a realistic case study.

Towards Negative Cycle Canceling in Wind Farm Cable Layout Optimization
Sascha Gritzbach, Torsten Ueckerdt, Dorothea Wagner, Franziska Wegner and Matthias Wolf (all KIT, DE)
In the Wind Farm Cabling Problem (WCP) the task is to design the internal cabling of a wind farm such that all power from the turbines can be transmitted to the substations and the costs for the cabling are minimized. Cables can be chosen from several available cable types, each of which has a thermal capacity and cost. Until now, solution approaches mainly use Mixed-integer Linear Programs (MILP) or metaheuristics. We present our current state of research on a fast heuristic specifically designed for WCP. We introduce an algorithm that iteratively improves a cable layout by finding and canceling negative cycles in a suitably defined network. Our simulations on publicly available benchmark sets show that the heuristic is not only fast but it tends to produce good results. Currently our algorithm gives better solutions on large wind farms compared to an MILP solver. However, on small to medium instances the solver performs better in terms of solution quality, which represents a starting point for future work.

CIMverter–a template-based flexibly extensible open-source converter from CIM to Modelica
Lukas Razik, Jan Dinkelbach, Markus Mirz and Antonello Monti (RWTH Aachen, DE)
Over the last decade, the Common Information Model (CIM), as specified by IEC 61970 / 61968, has become an important document format for the storage of power grid data. Its object-oriented design makes it easily maintainable and extensible for many use cases in the energy sector. As a result, more and more power grid analysis and simulation tools allow the import and export of CIM based power grid data. Unfortunately, many of them are proprietary and therefore not convenient in the research area since their component models and numerical back-ends often cannot be modified by the user. Thus, open-source alternatives are in demand, such as simulation environments based on the popular modeling language Modelica. Therefore, this paper presents our approach of a template based CIM to Modelica converter. The usage of templates makes it easily adaptable for the generation of Modelica system models targeting arbitrary Modelica libraries. The presented approach is implemented in an open-source project called CIMverter, evaluated on a real-world case with two Modelica power system libraries, and validated against a proprietary simulation tool.

Towards Model-driven CIM-based Data Exchange for DSOs
Dominik Ascher (TU Munich, DE) and Christoph Kondzialka (University of Applied Sciences Ulm, DE)
Smart energy systems (SES) promote the transformation of the distribution grid towards more sustainable operation and planning strategies, but also impose a set of considerable technological and political challenges. In this, distribution system operators (DSOs) are faced with the necessity of adapting their information system landscapes to enable the efficient utilization of information within their internal structures. In this work, we propose a model-based approach to derive an open middleware platform supporting the integration of existing system landscapes of DSOs. For this, we shortly describe a domain analysis on the DSO domain, which we use to derive the requirements of our platform. The platform is then implemented utilizing the Common Information Model (CIM) and open standards. Finally, we demonstrate the applicability of our approach within a small case study for a single use-case.

Session 5: Use Case-based Analysis

Session chair: Prof. Dr. Michael Sonnenschein (University of Oldenburg)

The Influence of Differential Privacy on Short Term Electric Load Forecasting
Günther Eibl (Salzburg University of Applied Sciences, AT), Kaibin Bao (KIT, DE), Philip-William Grassal, Daniel Bernau (both SAP, DE) and Hartmut Schmeck (KIT, DE)
There has been a large number of contributions on privacy-preserving smart metering with Differential Privacy, addressing questions from actual enforcement at the smart meter to billing at the energy provider. However, exploitation is mostly limited to application of cryptographic security means between smart meters and energy providers. We illustrate along the use case of privacy preserving load forecasting that Differential Privacy is indeed a valuable addition that unlocks novel information flows for optimization. We show that (i) there are large differences in utility along three selected forecasting methods, (ii) energy providers can enjoy good utility especially under the linear regression benchmark model, and (iii) households can participate in privacy preserving load forecasting with an individual membership inference risk < 60%, only 10% over random guessing. Waiting for the Sun – Can Temporal Flexibility in BEV Charging avoid Carbon Emissions?
Julian Huber and Christof Weinhardt (both KIT, DE)
Battery Electric Vehicles (BEVs) are claimed to foster climate-neutral energy and transportation systems. However, the use of the BEVs shifts emissions only geographically. In addition, the charging of BEVs can lead to new problems, such as congestions in the electricity grid. Smart charging algorithms can avoid some of these problems. To do this, however, the BEV user must actively decide to make her temporal flexibility available. Feedback on possible CO2 savings could be a nudge that encourages BEV users to provide more temporal flexibility. We compare different charging events and algorithms to determine the savings in CO2 emissions of BEV charging. We find that temporal exibility can on average save of CO2 emissions in the single-digit percentage range in Germany. The results can be used to implement a feedback nudge in smart charging systems to test its effectiveness on the provision of temporal charging flexibility of BEV users.

Explaining and Predicting Annual Electricity Demand of Enterprises – A Case Study from Switzerland
Carlo Stingl, Konstantin Hopf (both University of Bamberg, DE) and Thorsten Staake (University of Bamberg, DE, and ETH Zurich, CH)
In an attempt to channel sales activities, companies often focus on ‘high value targets’ that offer attractive prospective returns. In liberalized electricity markets, commercial customers with high electricity demand constitute such high value targets. The problem when acquiring new customers, however, is that the electricity consumption is not known to the sales organization in advance. This hinders the possibility to prioritize sales targets and thus increases the acquisition cost, reduces the competitiveness within the market and ultimately leads to higher cost for electricity customers. In this study, we investigate the annual electricity consumption of enterprises by means of a dataset with 1;810 company addresses in a typical town in Switzerland. We use the industry branch of the enterprises together with open big data (geographic information, online-content, social media data and governmental statistical data) to explain and predict the electricity consumption of such. Our linear regression analysis shows that information on the economic branches of the enterprises, basal area of buildings, number of opening hours and social media data can explain up to 19% of variance in electricity consumption. Economic trends (e.g., in labor market and turnover statistics) reflect changes in the electricity consumption in the investigated years 2010{2014 for several economic branches. We show, that the electricity consumption can be predicted better than a random predictor, however with a high uncertainty. Nevertheless, the open data sources can be used to identify a relevant group of companies with high consumption (more than 100;00 kWh per year) with good accuracy.