Time series data from building automation and control systems becomes available increasingly for analysis. This sort of data can be used for novel use cases such as fault detection or predictive maintenance. However, before such complex use case are implemented, it is important to have a base-line of consistency in the data otherwise the results will be unreliable. In a research project, we published simple data integrity checks that can find problems such as sensor drifts or misconfigurations of the automation system. In this talk, we will focus on the knowledge graph that we used to apply these checks automatically to buildings at scale. We will explain how we obtain semantic models of building automation systems, how these models are used to run the checks and the lessons we learned in the process.
Who might be interested in that (technical audience, management, data scientists, etc.)
All of the above
What would you like to know from other participants, what feedback are you looking for
Exchanging ideas around populating knowledge graphs with semantic data with minimal manual effort and while ensuring quality.
What is your background (name, affiliation, etc)
Maria Husmann, Team Lead Web of Things at Siemens Smart Infrastructure, Dr. in Computer Science from ETH Zurich.
Would you be open to have the session filmed & published later?
What would you like to talk about
Time series data from building automation and control systems becomes available increasingly for analysis. This sort of data can be used for novel use cases such as fault detection or predictive maintenance. However, before such complex use case are implemented, it is important to have a base-line of consistency in the data otherwise the results will be unreliable. In a research project, we published simple data integrity checks that can find problems such as sensor drifts or misconfigurations of the automation system. In this talk, we will focus on the knowledge graph that we used to apply these checks automatically to buildings at scale. We will explain how we obtain semantic models of building automation systems, how these models are used to run the checks and the lessons we learned in the process.
Who might be interested in that (technical audience, management, data scientists, etc.)
All of the above
What would you like to know from other participants, what feedback are you looking for
Exchanging ideas around populating knowledge graphs with semantic data with minimal manual effort and while ensuring quality.
What is your background (name, affiliation, etc)
Maria Husmann, Team Lead Web of Things at Siemens Smart Infrastructure, Dr. in Computer Science from ETH Zurich.
Would you be open to have the session filmed & published later?
Yes