Big Data for EnBop - Big Data analyses of automation data for the energetic operational optimization of existing buildings
The aim of the project is the development and evaluation of methods for the application of big-data analysis methods on building automation data for the identification of potentials for energetic operational optimization in existing buildings.
The innovative focus lies in the application of analysis methods for large and complex data sets (Big Data) to the enormous amount of operational data of building and component automation in modern buildings. Today, these are generally only used for the direct operational management of buildings and systems. Only a minimal amount of the data is used for visual inspections, alarms or basic analyses and reports. Most data is not stored or evaluated.
The approach pursued here is to systematically apply powerful big data methods, in particular through visualization/mapping and algorithms for data analysis, to historicized and real-time data from building automation systems and individual building services components such as heat pumps, boilers, ventilation units or pumps in order to analyse the potential of this data and develop utilization concepts.
Possible application scenarios include
- Component-specific analyses
- for operational optimization
- for a cost-optimized and preventive maintenance management
- to identify quality deficits of individual batches or installation companies
- System-specific analyses of data from buildings and facilities
- for identification and correction of operating errors
- for optimization of operating errors
- Quarter-specific analyses
- for the energetically and economically optimised operational management of plants in urban and grid contexts
In the project, typical data is analysed and evaluated both via simulations and in practice, in order to develop identification methods from these data for the applications described above. The methods are then tested in practice on real buildings and their facilities.
The intended methods should enable a comprehensive and largely automated identification of optimisation potentials in the building stock and thus, form the basis for a simpler, accelerated and more economical use of these potentials.