Rauno Jokelainen
CTO
The data presented here is collected through real-time measurements from locations where Truman Data’s standalone IoT devices are installed in collaboration with Truman Data’s Energy Coop PaaS. Real-time device data from energy assets is transmitted to the backend of Energy Coop, i.e., dedicated databases, through a secure virtual private network. Electricity consumption and the flexibility available for trading are forecasted using AI-based machine learning methods, and visualization is performed with cloud-native tools.
Learn More
Data collection devices from the platform, as well as those on the client side, are implemented as physical nodes within the Energy Coop PaaS virtual private network. Data connections utilize multi-layered encryption to ensure secure transfer of measurements from on-site devices to backend database servers. Data storage is structured with multiple isolated databases to further enhance data security and privacy.
Collected electricity consumption data is used for 48-hour forecasting with Prophet, leveraging its strength in time-series data featuring yearly, weekly, and daily seasonality, as well as holiday effects. Flexibility capacity is then calculated based on forecasted electricity consumption. This flexibility capacity is compared in real time with external market information using backend processes that run as cloud-native services on the Google Cloud Platform, with multiple layers of encryption.
Truman Data pilot in a Finnish city:
Real-time electricity consumption of connected buildings vs. demand response market data. Demonstrating how the Energy Coop solution empowers the city to generate new revenue by participating in flexibility markets, in addition to reducing energy consumption and generating savings.
Phase One of the project: Data from a sample of 15 connected municipal buildings.
Pilot phase of the 15 buildings based on 90 days of data
This first graph shows the real-time consumption and flexibility potential from the first 15 municipal buildings:
• Blue and green (MW): actual energy consumption.
• Green line: forecasted consumption.
• Purple line: flexibility capacity (can be up or down, but monetized flexibility is always positive).
• Consumption peaks reached 6 MW, with median consumption around 0.5 MW.
With the 15 buildings, there is a significant amount of flexibility that can be monetized.
Linear trend from project start to date: Consumption and available flexibility loads.
Sample of 15 connected buildings scaled to 70 city-owned buildings — showcasing the cloud’s full potential for steered demand response and savings.
Scaling up to all the city-owned 70 buildings based on 30 days of data
This graph simulates the entire building stock of the city’s schools, daycare, libraries, etc., the 70 properties:
• Blue (MW): total city-wide consumption, rising to over 20 MW.
• Purple (kW): available flexibility capacity, reaching up to 600 kW.
• The imbalance (flexibility) level target is around 100 kW.
At this scale, the city becomes an energy operator with its own “virtual power plant.”
Flexibility Revenue Opportunity in Riihimäki ›
48-hour real-time data from 15 connected buildings
48h live data of the pilot phase-one 15 buildings
The graph shows two days of live data from the 15-building pilot:
• Blue (MW): measured consumption.
• Green line: forecasted consumption.
• Purple (kW): flexibility capacity forecast, derived from measurements.
Live data demos how both consumption and flexibility can be measured and forecasted accurately, the platform provided to the city as a mechanism for it to participate in the flexibility market.
48-hour real-time data simulation for all 70 city-owned buildings
48h simulation of all the city-owned 70 buildings
Demonstrating 48 hours of the entire 70-building portfolio:
• Blue (MW): scaled-up consumption, exceeding 20 MW.
• Purple (kW): flexibility capacity across the city’s building stock.
Together, the city’s buildings are operating as a medium-sized virtual power plant (VPP), entirely based on controllable building demand.
Data drilldown 1: a single behind-the-meter energy asset in one building
Drill down on a single building
This graph shows device-level control from a single location (smaller school/daycare):
• Blue (kW): energy consumption.
• Sharp spikes and drops represent devices being switched on and off.
• Downward drops show controlled load reductions, releasing flexibility to the market.
A single building can generate tradable flexibility can be aggregated become part of the City’s VPP’s energy portfolio.
Prediction model for batching large demand response from small sources.
Data drilldown 2: Buy and sell orders with matched sales
Market trading 15 days of data
The data graph shows how flexibility translates into actual market revenues:
• Top (EUR/MWh): order book with sell and buy orders. When the market price exceeds the customer’s threshold of their target price, EUR 100/MWh, flexibility capacity generates sell orders.
• Bottom (purple bars): matched deals (when buy and sell orders meet), showing actual traded volume.
• In 15 days, the city generates EUR 7,875 in new revenue, projected to be EUR +/- 189,000 annually.
The system automatically matches flexibility supply with demand on the market, creating a steady revenue stream for the city.
Energy flexibility activated at a predefined market price threshold ›
