"This site requires JavaScript to work correctly"
Header minimieren

Jakob Kasbauer, M.Sc.

Mobile and Embedded Systems (Universität Passau) Forschungsgruppe: Applied Artificial Intelligence Datenanalyse und Machine Learning

Zentrum für angewandte Forschung

Technologie Campus Grafenau

Wissenschaftlicher Mitarbeiter

Grafenau

08552/975620-34

Sortierung:
Beitrag in Sammelwerk/Tagungsband

  • Sebastian Wilhelm
  • Dietmar Jakob
  • Jakob Kasbauer
  • Diane Ahrens

GeLaP: German Labeled Dataset for Power Consumption

In: Proceedings of Sixth International Congress on Information and Communication Technology. null (Lecture Notes in Networks and Systems) vol. 235 pg. 21-33.

Singapore

  • (2022)

DOI: 10.1007/978-981-16-2377-6_5

Due to the increasing spread of smart meters, numerous researchers are currently working on disaggregating the power consumption data. This procedure is commonly known as Non-Intrusive Load Monitoring (NILM). However, most approaches to energy disaggregation first require a labeled dataset to train these algorithms. In this paper, we present a new labeled power consumption dataset that was collected in 20 private households in Germany between September 2019 and July 2020. For this purpose, the total power consumption of each household was measured with a commercial available smart meter and the individual consumption data of 10 selected household appliances were collected.
  • TC Grafenau
  • DIGITAL
Zeitschriftenartikel

  • Sebastian Wilhelm
  • Jakob Kasbauer

Exploiting Smart Meter Power Consumption Measurements for Human Activity Recognition (HAR) with a Motif-Detection-Based Non-Intrusive Load Monitoring (NILM) Approach

In: Sensors vol. 21 pg. 8036.

  • (2021)

DOI: 10.3390/s21238036

Numerous approaches exist for disaggregating power consumption data, referred to as non-intrusive load monitoring (NILM). Whereas NILM is primarily used for energy monitoring, we intend to disaggregate a household’s power consumption to detect human activity in the residence. Therefore, this paper presents a novel approach for NILM, which uses pattern recognition on the raw power waveform of the smart meter measurements to recognize individual household appliance actions. The presented NILM approach is capable of (near) real-time appliance action detection in a streaming setting, using edge computing. It is unique in our approach that we quantify the disaggregating uncertainty using continuous pattern correlation instead of binary device activity states. Further, we outline using the disaggregated appliance activity data for human activity recognition (HAR). To evaluate our approach, we use a dataset collected from actual households. We show that the developed NILM approach works, and the disaggregation quality depends on the pattern selection and the appliance type. In summary, we demonstrate that it is possible to detect human activity within the residence using a motif-detection-based NILM approach applied to smart meter measurements.
  • TC Grafenau
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • Sebastian Wilhelm
  • Dietmar Jakob
  • Jakob Kasbauer
  • Melanie Dietmeier
  • A. Gerl
  • Benedikt Elser
  • Diane Ahrens

Organizational, Technical, Ethical and Legal Requirements of Capturing Household Electricity Data for Use as an AAL System

In: Proceedings of the Fifth International Congress on Information and Communication Technology (ICPCCI 2019) [20-21 February 2020; London, UK]. null (Advances in Intelligent Systems and Computing)

Singapore

  • (2020)

DOI: 10.1007/978-981-15-5856-6_38

Due to demographic change elderly care is one of the major challenges for society in near future fostering new services to support and enhance the life quality of the elderly generation. A particular aspect is the desire to live in one’s homes instead of hospitals and retirement homes as long as possible. Therefore it is essential to monitor the health status i.e. the activity of the individual. In our data-driven society data is collected at an increasing rate enabling personalized services for our daily life using machine-learning and data mining technologies. However the lack of labeled datasets from a realistic environment hampers research for training and evaluating algorithms. In the project BLADL we use data mining technologies to gauge the health status of elderly people. Within this work we discuss the challenges and caveats both from a technical and ethical perspectives to create such a dataset.
  • TC Grafenau
  • Angewandte Wirtschaftswissenschaften
  • DIGITAL