Could you please introduce yourself, your company, and your role in the STREAM project? 

Within the STREAM EU project, whose scope is to explore possibilities to provide flexibility services on the low voltage grid, OptiWatti is prototyping an innovative mechanism for incentive-driven flexible usage of active power to help stabilizing and optimising the grid. This prototype explores the possibility to use a pre-defined part of the load used for heating the buildings as a Frequency Containment Reserve or FCR. 

FCRs are active power reserves that are automatically controlled based on the grid’s frequency deviation. Their purpose is to contain the frequency variations during normal operation and during disturbances. 

Our pilot site includes 7 residential buildings and 3 buildings that are part of a holiday resort. When defining the use cases and the requirements, we have been considering from day one the possibility to scale the solution to the full OptiWatti customer base. 

Can you provide an overview of the Finnish pilot site, including its current scope and main focus? 

Within the STREAM EU project, whose scope is to explore possibilities to provide flexibility services on the low voltage grid, OptiWatti is prototyping an innovative mechanism for incentive-driven flexible usage of active power to help stabilizing and optimising the grid. This prototype explores the possibility to use a pre-defined part of the load used for heating the buildings as a Frequency Containment Reserve or FCR. 

FCRs are active power reserves that are automatically controlled based on the grid’s frequency deviation. Their purpose is to contain the frequency variations during normal operation and during disturbances. 

Our pilot site includes 7 residential buildings and 3 buildings that are part of a holiday resort. When defining the use cases and the requirements, we have been considering from day one the possibility to scale the solution to the full OptiWatti customer base. 

What are the primary goals of the Finnish pilot site?

We have few important goals. The first one is to study the technical feasibility of such a solution, by taking into consideration the challenging requirements set by the TSO on the Aggregator (that is, us) in terms of accuracy of the amount of the sold flexible load in the bid, minimum amount of offered flexible load, activation time, etc … This includes also defining how to integrate the demand-response functionality that we are prototyping in the STREAM project with the existing implicit flexibility, that is, consumer reaction to price variations (which we call spot-price based control). Those two pieces of functionality could have an opposite effect on the same load since the first is connected to grid frequency and the second one to the energy price. So, before doing any bid to the TSO in the demand-response, we have to take into account customer settings and the energy spot prices for the next coming day. 

At the same time, we study the business feasibility of the solution with particular attention to the costs and benefits for the customer when his/her assets are part of an FCR. Finally, we are investigating the impact on the tenet’s satisfaction and we collect periodically feedback from our pilot customers. 

How will the flexibility from heating be utilized for services like FCR for TSOs? 

The classic way to balance production and consumption of electric power is to modulate the production of energy. What we are doing here is to operate on the loads. In our case a decrease in grid’s frequency will cause a decrease in load while an increase in grid’s’ frequency will cause an increase in load. 

The prototype includes 3 flexibility services: FCR-N (where N stays for normal operations), FCR-D-upwards and FCR-D-downwards (where D stays for disturbances). Each of those services have a defined standard frequency range. 

Our role as an aggregator will be to estimate how much flexible-power we can promise to Fingrid, the Finnish TSO. Then, we create a bid for each hour of the next day and we send it to the TSO. Once the TSO accepts our bid, we start to monitor the grid’s frequency. 

If we take as an example the FCR-D-upward service, if we notice that the frequency goes below 49.9 Hz, we start to cut heating load. The amount of load we cut is proportional to the frequency deviation. For example, a frequency of 49.7 Hz causes cutting half of the load that we agreed in the bid. A frequency of 49.5 Hz would cause cutting the full load in the bid. With FCR-D-downward service we monitor if the frequency goes too high and, in that case, we add heating load rather than cutting it. 

What is the impact you foresee on the building’s occupants and what measures are you taking in order to solve any potential issue?

According to statistics we have received from Fingrid, the average duration of each frequency disturbance lasts less than a minute and during this time there will be no visible temperature deviation in the rooms. However, we expect an average of 100 frequency deviations per day. Therefore, if we would use the same assets for all the FCR activations of the day, the noise created by the relays which switch the heating load might be an issue causing tenant’s discomfort in certain cases. 

One way to mitigate this effect is to add customer settings that allow to decide at which time of the day the building assets can be used for demand-response, for example forbidding the usage during the night hours. 

A positive effect on the customers will be that they will be able to get a share of the revenues that the Aggregator gets from the TSO as a result of the bid process for having offered flexible load. 

What have been the main challenges so far?

One example is the accuracy required for the reported activated power. Ideally, we would meter this power. However, this would require installing metering devices in every room and this would be too expensive and would kill the business case. So, we found a solution which consists in calculating the consumed energy in each room by using data about the activities of the relays controlling the load and the value of the installed power in the heating devices in the rooms. Also, we set the thermostats in the actual heating devices to a high value in order to prevent situations when the relay switches-on the load but the device is not consuming power because the thermostat is regulated to a too low value. To prove that our calculated power corresponds to real consumed power, we measure the energy consumed by the whole apartment with a single metering device and a sampling interval of few seconds to verify that the actual activated power is corresponding to the calculated power that we have used in the bid to the TSO. Finally, we agree with the TSO to use a capacity reduction factor (which will be calculated based on the outcome of the previous step), which means that we will have to activate more flexible capacity than what we sell. 

What have been the main lessons learned?

Organizing regular workshops with the pilot customers has helped to keep them engaged and to receive precious feedback already in the early phases of the project. Cooperation with TSO as been at least as important for proceeding successfully with the project. 

In our preliminary study we found that certain types of loads, like for example hot water storage or car heaters/charging, are not suitable assets for FCR because the level of control the Aggregator has currently on those assets is not enough. Therefore, we are using only electric radiators and electric floor heating as FCR assets in at this stage. 

A third result of the study so far is that customers want to be in full control of or at least informed about the usage of their devices. They want to know when their assets are used in a FCR group. Therefore, we are planning modifications to OptiWatti user interface to satisfy this requirement. 

What is the project status and what are the next steps for STREAM? 

At the beginning of March at the 6th STREAM general meeting, we demonstrated our capability to run FCR-D-up service with a simulated interface towards the TSO and with simulated load. 

The next steps will be to complete this functionality by removing the above-mentioned limitations and by implementing additional requirements coming from the TSO like normalization of the load. Then, by using the basic building blocks created for FCR-D-upwards, we will implement also other services like FCR-D-downwards. Also, we will add the needed customer settings and we will do modifications to the plots shown in the user interface. The development project will last till June 2026.