Energy saving, CO² reduction & optimisation of the indoor climate of a court house building in Tallinn

The R8 Autopilot took over the control of Air Handling Units (AHUs) and room controllers in the building and thereby optimised the heating circuit control of ventilation. Since then, the climate changed gradually.

General Background

The pilot project for the Lubja 4 office building started in mid-August 2019 and consisted of a total of three main phases: integration, diagnostics and system management (autopilot). During the first phase, the building automation system was integrated to R8tech software and data collection started from the building's existing automation system server (BMS).As part of the second phase, diagnostics began work and at the end of September we presented a building’s diagnostics report. The purpose of the report was to provide an overview of the technical condition of the building by means of faults and other observations detected by automatic fault detection. As part of the third phase, we launched the Autopilot that involved the continuous model-based optimization of the building's HVAC system settings. The Autopilot in this building was fully implemented between 01.11.19 - 29.02.20 (4 months).

Technical Description

There are 2 chillers in the building that supply ventilation and room cooling devices with cooling energy. One chiller that is connected with room cooling devices has a free cooling capacity as well. The heating source in the building is district heating, which supplies four different heating circuits: domestic hot water, radiator heating, floor heating and ventilation heating. The building is served by a total of 16 ventilation units and all of them have a heat recovery system (rotary or glycol).

The rooms have room controllers that control VAV damper positions and room heating/cooling devices. The heaters in the rooms are mostly radiators, but there are some areas in the building where floor heating is being used. Depending on the room, the room cooling devices are either active chilled beams or fancoils. In addition, the room has presence-based lighting control.
 

R8 Autopilot

At first, rule based predictive control was launched in the building. Autopilot took over control of Air Handling Units (AHUs) and room controllers. The control parameters were adjusted based on the indoor climate preferences of tenants. In the following weeks, control of heating circuit of ventilation was started, since then climate changed gradually.

Heating impact of HVAC components was calculated for each area individually to provide only necessary heating during occupancy hours. The advance heating optimization was accomplished by interconnecting with the building ecosystem (weather effect, heating capacity and its impact, tenant interactions etc.). Meanwhile, heating circuit models were trained to improve the accuracy of adjustments.

In total, Autopilot changed 1,279 different setpoints during the entire control period (including 1,248 setpoints related to premises). A total of 682,000 changes were made, of which 640,000 were related to room setpoints and 27,100 changes were made to ventilation air temperature. The total number of setpoints, which were changed more than 100 times, was 907.

During the period of autopilot implementation, the average monthly electricity savings were 16.3% and for heat energy 40.4%. In four months, a total of 119.8 MWh of electricity and 660.0 MWh of heat were saved. In total, the amount of saved energy was 779.8 MWh over four months and the financial savings were € 41400.40.