Venue: Aalborg University Copenhagen, Frederikskaj 10A, Building D, 3rd floor, room 3.152.

Registration

Workshop participation is free, but registration is required.

About the workshop

Recent projects in the field of district heating systems and information technology have focused on analyzing data from District Heating Substations (DHS) for several purposes:

• Fault detection and prediction with confidence in time series data (e.g., DAD project)
• Anomaly detection prediction and classification using deep learning (e.g., DAD and TERMO projects)
• Identifying energy consumption patterns among buildings (e.g., DAD, SeMI and TEMPO projects)
• Integrating fault detection models into decision support for performance monitoring (e.g., DAD, BigData@BTH and SeMI projects)
• How to construct self-monitoring systems that use joint-human machine learning to adapt to specific domains, by taking advantage of groups of peers, and ubiquitous streams of data (e.g., SeMI project)

This workshop aims to narrow the gap between industrial and academic efforts toward understanding the performance of a DHS. The workshop address the following key question:

“How to determine when performance degradation becomes a fault?”

Presentations from industrial experts shall highlights processes employed in fault identification, extraction, and treatment, aimed at ensuring the overall performance of the substation. These processes shall be concretize with real world examples from different industries.

Presentations from academia shall demonstrate how to employ DHS data to detect and isolate unexpected changes in substations, faulty substations, or substations with unacceptable poor performance. The demonstrations shall involve concrete use-cases from different research projects.

Expected outcomes from the workshop

1. Discussions aimed identifying key industrial demands for improving processes related to fault identification, extraction, and treatment in a DHS.
2. Discussions aimed at identifying key performance measurements or ranking of different substations performances.
3. Discussions aimed at initiating a list of relevant problems to solve in order to achieve meaningful industrial progress with respect to substation performance monitoring.
   
   

Organisers of the workshop

Borås University and Noda Intelligent Systems AB are the lead organisers of the workshop with support from Halmstad University, Blekinge Institute of Technology, Lund University, Aalborg University and Öresundskraft. The workshop is organized alongside the 5^th International Conference on Smart Energy Systems.

Links to research projects related to the workshop:

• DAD- Data Analytics for Fault Detection in District Heating (DAD)
• SeM (external link)
• BigData@BTH - Scalable resource-efficient systems for big data analytics (external link)
• TEMPO (external link)

For more information about the workshop, contact: gideon.mbiydzenyuy@hb.se, jens.brage@noda.se