2021

 back to the year overview

• Benjamin Völker, Philipp M Scholl, Bernd Becker
  A Feature and Classifier Study for Appliance Event Classification
  2021 3rd EAI International Conference on Sustainable Energy for Smart Cities - EAI SESC 2021 Springer
  » show abstract« hide abstract
  Abstract

  The shift towards advanced electricity metering infrastructure gained traction because of several smart meter roll-outs during the last decade. This increased the interest in Non-Intrusive Load Monitoring. Nevertheless, adoption is low, not least because the algorithms cannot simply be integrated into the existing smart meters due to the resource constraints of the embedded systems. We evaluated 27 features and four classifiers regarding their suitability for event-based NILM in a standalone and combined feature analysis. Active power was found to be the best scalar and Waveform Approximation the best multidimensional feature. We propose the feature set [P, cosPhi, TRI, WFA] in combination with a Random Forest classifier. Together, these lead to F1-scores of up to 0.98 on average across four publicly available datasets. Still, feature extraction and classification remains computationally lightweight and allows processing on resource constrained embedded systems.
• Benjamin Völker
  Acquiring, congregating, and processing high-frequency electricity data in distributed environments
  2021
  » show abstract« hide abstract
  Abstract

  The energy consumption of a home depends on the behavior of its inhabitants offering a promising energy saving potential. However, this potential can only be unfolded to full extent if the consumption of each individual appliance is know. Non-Intrusive Load Monitoring (NILM) offers a retrospective way to get individual appliance consumption data. If such data are combined with eco-feedback techniques it can help to better understand a user’s electricity usage to ultimately save energy. Researching NILM algorithms, and in particular, the development of the underlying supervised machine learning techniques, requires adequate datasets with corresponding ground truth data and methodologies to create more labeled data when needed. Adding detailed labels to such datasets is a time-consuming and error-prone process. Deploying a supervised NILM system typically requires a dedicated system training procedure hampering their widespread adoption. The thesis at hand presents several strategies to address these challenges in order to improve the adoption of NILM. In particular, a set of requirements is presented for acquiring and congregating high-frequency electrical measurements in distributed environments. These are handled by a novel recording framework comprised of a central recording director and two prototype Data Acquisition Systems (DAQs), one for aggregated and one for plug-level data. The developed methodologies allow the DAQs to deliver highly accurate and time-synchronized data while using rather inexpensive components. To add precise and descriptive labels to such data, a semi-automatic labeling method is developed and evaluated on two publicly available datasets. The method improves the labeling efficiency up to 74% and has been integrated into a novel labeling tool implemented as a web-application. The framework and labeling tool have been used to collect and label the Fully-labeled hIgh-fRequency Electricity Disaggregation (FIRED) dataset. It contains 101 days of 8 kHz aggregated current and voltage measurements of the 3-phase electricity supply of a typical residential apartment in Germany. The data also includes synchronized 2 kHz plug-level readings of 21 individual appliances, other environmental sensor measurements, and descriptive event labels of all appliances, resulting in a complete and versatile residential electricity dataset. Furthermore, several domain specific features and classifiers are evaluated regarding their suitability for (event-based) NILM targeted for resource constrained systems. Data cleaning methods are evaluated which remove the steady-state energy consumption of other appliances from the aggregated data of a given appliance event. As plug-level data delivers less noisy individual appliance data, it is shown that the inclusion of such data during training results in a performance gain for appliance classification algorithms. Finally, a novel supervised NILM system is proposed and evaluated which uses a combination of aggregated and individual appliance data to improve and aid the training process while only requiring minimal user interaction.
• Benjamin Völker, Marc Pfeifer, Florian Wolling, Philipp M. Scholl, Bernd Becker
  Introducing MILM - A Hybrid Minimal-Intrusive Load Monitoring Approach
  2021 The Twelfth ACM International Conference on Future Energy Systems (e-Energy ’21) ACM
  » show abstract« hide abstract
  Abstract

  The shift towards an advanced electricity metering infrastructure has gained traction because of several smart meter roll-outs. This accelerated research in Non-Intrusive Load Monitoring techniques. These techniques highly benefit from the temporal resolution improvements achieved by smart meters. Nevertheless, industrial adoption is low, not least because the achieved disaggregation performance is rather poor for unsupervised approaches. This work sketches a way to utilize intrusive sensors in combination with a standard NILM system to enhance training and maximize overall system's performance while minimizing the number of required intrusive sensors.
• Marc Pfeifer, Benjamin Völker, Sebastian Böttcher, Sven Köhler, Philipp M Scholl
  Teaching Embedded Systems by Constructing an Escape Room
  2021 In Proceedings of the 52nd ACM Technical Symposium on Computer Science Education (SIGCSE ’21), Virtual Event, New York, USA ACM
  » show abstract« hide abstract
  Abstract

  Embedded systems form the basis of many of today's consumer products and industrial applications - and they are increasingly connected. To teach this topic, we created a course with the overarching goal of designing and constructing an automated escape room. This provided the motivation for the students to learn the engineering and soft skills required for building networked embedded systems. The game was open for faculty members and friends of the students after course completion. By splitting the building process into multiple tasks, such as individual puzzles, the presented concept encourages inter- and intra-group work, including conceptualizing, designing and developing reliable, connected embedded systems. In this paper we first present the motivation, context, and pedagogical approach of the course. We then describe the course structure and conclude with experiences from constructing an escape room as a group.
• Benjamin Völker, Andreas Reinhardt, Anthony Faustine, Lucas Pereira
  Watt’s up at Home? Smart Meter Data Analytics from a Consumer-Centric Perspective
  2021 Energies, volume: 14, issue: 3
  » show abstract« hide abstract
  Abstract

  The key advantage of smart meters over traditional metering devices is their ability to transfer consumption information to remote data processing systems. Besides enabling the automated collection of a customer’s electricity consumption for billing purposes, the data collected by these devices makes the realization of many novel use cases possible. However, the large majority of such services are tailored to improve the power grid’s operation as a whole. For example, forecasts of household energy consumption or photovoltaic production allow for improved power plant generation scheduling. Similarly, the detection of anomalous consumption patterns can indicate electricity theft and serve as a trigger for corresponding investigations. Even though customers can directly influence their electrical energy consumption, the range of use cases to the users’ benefit remains much smaller than those that benefit the grid in general. In this work, we thus review the range of services tailored to the needs of end-customers. By briefly discussing their technological foundations and their potential impact on future developments, we highlight the great potentials of utilizing smart meter data from a user-centric perspective. Several open research challenges in this domain, arising from the shortcomings of state-of-the-art data communication and processing methods, are furthermore given. We expect their investigation to lead to significant advancements in data processing services and ultimately raise the customer experience of operating smart meters.

2020

 back to the year overview

• Benjamin Völker, Marc Pfeifer, Philipp M Scholl, Bernd Becker
  A Framework to Generate and Label Datasets for Non-Intrusive Load Monitoring
  2020 Energies
  » show abstract« hide abstract
  Abstract

  In order to reduce the electricity consumption in our homes, a first step is to make the user aware of it. Raising such awareness, however, demands to pinpoint users of specific appliances that unnecessarily consume electricity. A retrofittable and scalable way to provide appliance-specific consumption is provided by Non-Intrusive Load Monitoring methods. These methods use a single electricity meter to record the aggregated consumption of all appliances and disaggregate it into the consumption of each individual appliance using advanced algorithms usually utilizing machine-learning approaches. Since these approaches are often supervised, labelled ground-truth data need to be collected in advance. Labeling on-phases of devices is already a tedious process, but, if further information about internal device states is required (e.g., intensity of an HVAC), manual post-processing quickly becomes infeasible. We propose a novel data collection and labeling framework for Non-Intrusive Load Monitoring. The framework is comprised of the hardware and software required to record and (semi-automatically) label the data. The hardware setup includes a smart-meter device to record aggregated consumption data and multiple socket meters to record appliance level data. Labeling is performed in a semi-automatic post-processing step guided by a graphical user interface, which reduced the labeling effort by 72% compared to a manual approach. We evaluated our framework and present the FIRED dataset. The dataset features uninterrupted, time synced aggregated, and individual device voltage and current waveforms with distinct state transition labels for a total of 101 days.
• Benjamin Völker, Marc Pfeifer, Philipp M Scholl, Bernd Becker
  Annoticity: A Smart Annotation Tool and Data Browser for Electricity Datasets
  2020 5th International Workshop on Non-Intrusive Load Monitoring ACM
  » show abstract« hide abstract
  Abstract

  The growing request for eco-feedback and smart living concepts accelerated the development of Non-Intrusive Load Monitoring (NILM) algorithms during the last decade. Comparing and evaluating these algorithms still remains challenging due to the absence of a common benchmark datasets, and missing best practises for their application. Despite the fact that multiple datasets were recorded for the purpose of comparing NILM algorithms, many researchers still have to record their own dataset in order to meet the requirements of their specific application. Adding ground truth labels to these datasets is a cumbersome and time consuming process as it requires an expert to visually inspect all the data manually. Therefore, we propose the Annoticity inspection and labeling tool which simplifies the process of visualizing and labeling of electricity data. We use an event detector based on the log likelihood ratio test which achieved an F1 score of 90.07% in our experiments. Preliminary results indicate that the effort of generating event labels is reduced by 80.35% using our tool.
• Benjamin Völker, Marc Pfeifer, Philipp M Scholl, Bernd Becker
  FIRED: A Fully-labeled hIgh-fRequency Electricity Disaggregation Dataset
  2020 The 7th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation Virtual Event Japan November, 2020 ACM
  » show abstract« hide abstract
  Abstract

  As more and more homes are equipped with smart electricity meters, home owners gain the ability to monitor their total electricity consumption on a daily or hourly basis. Techniques such as load forecasting, load disaggregation, and activity recognition try to provide even better insights into our electricity consumption, highlight saving potential or improve our daily living. To develop and evaluate these techniques, publicly available datasets are used. We identified a lack of high frequency fully labeled electricity datasets in the residential domain and present the FIRED dataset. It contains 52 of 8kHz aggregated current and voltage readings of the 3-phase supply of a typical residential apartment in Germany. The dataset also contains synchronized ground truth data as 2kHz readings of 21 individual appliances, as well as room temperature readings and fully labeled state changes of the lighting system, resulting in a complete and versatile residential electricity dataset.

2019

 back to the year overview

• Benjamin Völker, Marc Pfeifer, Philipp M Scholl, Bernd Becker
  A Versatile High Frequency Electricity Monitoring Framework for Our Future Connected Home
  2019 EAI International Conference on Sustainable Energy for Smart Cities, Braga Portugal Springer, pages: 221 - 231
  » show abstract« hide abstract
  Abstract

  In our homes a lot of devices are powered by electricity with- out us knowing the specific amount. As electricity production has a large, negative environmental impact, we should be more aware about how de- vices consume power and how we can adapt our daily routine to decrease our electricity requirements. Methods such as Non-Intrusive Load Mon- itoring (NILM) can provide the user with precise device level electricity data by measuring at a single point in a houses’ electricity network. However, the time resolution of most off-the-shelf power meters is not sufficient for NILM or the meters are locked down for security reasons. Therefore, we have developed our own versatile energy metering frame- work which consists of a high frequency electricity metering device, a versatile backend for data processing and a webapp for data visualiza- tion. The developed hardware is capable of sampling up to 32 kHz, while the software framework allows to extract other power related metrics such as harmonic content. The system’s application ranges from pro- viding transparent electricity usage to the user up to generating load forecasts with fine granularity.
• Benjamin Völker, Philipp M Scholl, Bernd Becker
  Semi-Automatic Generation and Labeling of Training Data for Non-intrusive Load Monitoring
  2019 e-Energy '19 Proceedings of the Tenth ACM International Conference on Future Energy Systems ACM New York, NY, USA ©2019, volume: 10, pages: 17 - 23
  » show abstract« hide abstract
  Abstract

  User awareness is one of the main drivers to reduce unnecessary energy consumption in our homes. This awareness, however, requires individual energy data of the devices we own. A retrofittable way to get this data is to use Non-Intrusive Load Monitoring methods. Most of these methods are supervised and require to collect labeled ground truth data in advance. Labeling on-phases of devices is already a tedious process, but if further information about internal device states are required (e.g. intensity of an HVAC), manual labeling methods are infeasible. We propose a novel data collection and labeling method for Non-Intrusive Load Monitoring. This method uses intrusive sensors directly connected to the monitored devices. A post-processing step classifies the connected devices into four categories and exposes internal state sequences in a semi-automatic way. We evaluated our labeling method with a sample dataset comparing the amount of recognized events, states and classified device category. The event detector achieved a total F1 score of 86.52 % for devices which show distinct states in its power signal. Using our framework, the overall labeling effort is cut by more than half (42%).
• Phillip M Scholl, Benjamin Völker, Bernd Becker, Kristof Van Laerhoven
  A multi-media exchange format for time-series dataset curation
  In: Human Activity Sensing: Corpus and Applications
  2019, Springer International Publishing, pages: 111 - 119, ISBN: 9783030130015
  » show abstract« hide abstract
  Abstract

  can be applied to optimize storage and transmission overhead.

2018

 back to the year overview

• Benjamin Völker, Philipp M Scholl, Bernd Becker
  Monitoring - A Hybrid Approach
  2018 e-Energy '18 Proceedings of the Ninth International Conference on Future Energy Systems/Karlsruhe, Germany , pages: 436 - 438
  » show abstract« hide abstract
  Abstract

  research directions for hybrid load monitoring systems

2017

 back to the year overview

• Tobias Schubert, Benjamin Völker, Marc Pfeifer, Bernd Becker
  The Smart MiniFab: An Industrial IoT Demonstrator Anywhere at Any Time
  2017 Smart Education and e-Learning 2017, Vilamoura/Portugal, KES International Springer International Publishing, pages: 253 - 262

2016

 back to the year overview

• Benjamin Völker, Tobias Schubert, Bernd Becker
  iHouse: A Voice-Controlled, Centralized, Retrospective Smart Home
  2016 7th EAI International Conference on Sensor Systems and Software
  » show abstract« hide abstract
  Abstract

  Speech recognition in smart home systems has become pop- ular in both, research and consumer areas. This paper introduces an in- novative concept for a modular, customizable, and voice-controlled smart home system. The system combines the advantages of distributed and centralized processing to enable a secure as well as highly modular plat- form and allows to add existing non-smart components retrospectively into the smart environment. To interact with the system in the most com- fortable way - and in particular without additional devices like smart- phones - voice-controlling was added as the means of choice. The task of speech recognition is partitioned into decentral Wake-Up-Word (WUW) recognition and central continuous speech recognition to enable flexibil- ity while maintaining security. This is achieved utilizing a novel WUW algorithm suitable to be executed on small microcontrollers which uses Mel Frequency Cepstral Coefficients as well as Dynamic Time Warping. A high rejection rate up to 99.93% was achieved, justifying the use of the algorithm as a voice trigger in the developed smart home system.