Autonomous Management System Developed for Building and District Levels
The system works by managing energy flows and also by predicting and mastering energy consumption and production. It was intended that individual businesses be able to prioritise reduction of costs CO2 emissions or various other options and combinations. Key to the system is the District Energy Management and Information System (DEMIS). The unit receives information from and interacts with other control systems to optimise costs and energy consumption.
After planning the initial architecture and setting objectives, the team implemented a prototype system at three demonstration sites. Using tools developed within the project, researchers simulated and tested various simple and complex models. The team developed algorithms that enable predictive control of buildings or a district. Testing showed that energy can be optimised within a district. Campuses were successfully managed using centralised controllers, whereas the team used distributed optimisation to control districts containing multiple organisations. For the latter method, researchers tested primal and dual decomposition. Results indicated substantial savings at the district level. At the building level, savings ranged from 10 to 40 %, with an average of 28 %. The team’s algorithms for district heating yielded a 20 % reduction in gas consumption. Similarly, the project’s algorithms lowered costs in multiple-organisation districts by between 10 and 20 %.