The Linear research project examined various methods to tailor a household's energy consumption in function of the available wind and solar energy. These methods included new technologies and user interfaces.



Closed project



The major research questions were:

- How does a change in behavior benefit both households and the industry?
- How are the costs and benefits distributed among the parties involved?
- Which solutions result in sufficient motivation and convenience to prompt a change in behavior?
- To what extent are households willing and able to adjust their behavior?

Local Intelligent Networks and Energy Active Regions

Local Intelligent Networks and Energy Active Regions, or in short 'Linear', was a research project bringing together twenty different partners, including KU Leuven, VITO, iMinds, imec, industrial associates and the Government of Flanders. Linear investigated the best and easiest ways to adapt residential energy consumption based on the availability of solar and wind energy with the help of 250 test households.

The ambitious pilot project was widely acknowledged, receiving the European EEGI-label and the global ISGAN Award. Various products and services are now further researched by the Linear partners.



In our electricity grid, the production of electricity constantly follows demand, that's the basic principle of the grid. This, however, becomes ever more difficult to maintain. The capacity of classic production decreases: energy providers close down older and inefficient power plants and nuclear production is gradually reduced. At the same time, we install extra production, most often in the shape of wind or solar energy, but those depend on the weather, not demand of energy.

This sometimes results in energy surpluses, such as for example during the sunny Pentecost weekend of 2012 and the sunny Easter Monday of 2013, when so much excessive power was produced that the grid almost failed. Similar reasoning applies for a higher chance of energy shortage, during cold, grey days when there is no wind available and demand is high.  

Demand response, a solution?

These changes in the production model can partly be solved by turning the logic upside down: by aligning the energy use and offer. Within Linear, some twenty partners investigated how the energy use of 250 test case families can best and easiest be adapted at home to the availability of solar and wind energy.

Incentive models

Linear used two incentive models and four business cases to investigate how families and suppliers or grid operators can better align the energy consumption and production. Encouraging families to shift their energy consumption, can be done by providing a financial incentive (tariff control), or by automatically controlling their appliances (automatic control) and reward the families for the flexibility offered.

Automatic demand response

For three quarters of the participants an automated system was installed in which Linear controlled the appliances. Linear did this when needed, when energy surpluses or imbalances occurred in the grid or when the price was low. While doing so, the comfort of the family was constantly guaranteed.

To illustrate the concept we provide an example: a test family wants to have their dishes cleaned by 6 PM at night. They switch their dishwasher on at 8 AM in the morning and install 6 PM as ultimate deadline. The appliance requires 2 hours to clean the dishes, so needs to start at 4 PM latest. This way a time frame is enabled between 8 AM and 4 PM. During these 8 hours of flexibility Linear can remotely start the appliance. If the machine hasn’t started by 4 PM, it will automatically start so that the dishes will be cleaned by 6 PM. This is an example of automatic demand response in Linear.

This, however, only works if the energy supplier is able to communicate with the consumer. This therefore requires smart appliances. When and based on which criteria the appliance should start performing its task, was investigated by means of four business cases.

Linear combined automatic demand response with tariff control: the appliances would be switched on when better prices were available. The ultimate aim is that the consumer also benefits from automatic demand response.

Variable tariff control

The simplest and oldest example of tarrif control in the energy sector is the day/night tariff. To not having to adapt the production during the night (when demand is low) and day (when demand is high), energy providers use lower prices at night. This way they stimulate their customers to shift their energy use as much as possible

Until a few years ago, classic energy sources led to a stable energy production, with often only differences between summer and winter. With the rise of renewable energy sources, however, production has become variable, with highly unpredictable energy surpluses as a result. During the day in a weekend for example, when the sun shines bright, or at night when there is a lot of wind available,… Similar to night tariffs, the price can be lowered in these cases, but are not as predictable as day/night tariffs.

Former European projects already demonstrated that unpredictable variable tariffs will only little change the consumption pattern. That’s why Linear only used this concept (without automatic demand response) for a quarter of the participants. By means of a display, these families were given insight into 6 tariff blocks per day of which prices differed according to the sun, wind, demand and offer predicted. Following the time of use concept the day was divided into six fixed timeframes. On a daily basis the energy tariff for the next day was determined, taking into consideration the production of solar and wind energy and the expected consumption. The idea was that participants would schedule their electricity use in cheaper time frames so that economic advantage could be achieved. Additionally, the total energy consumption could easily be monitored and controlled.

Linear Variabele tariefstructuur

Case Studies

Next to the user interaction, Linear also evaluated four business cases enabling energy suppliers to integrate renewable energy sources into the grid.

As we gradually make the shift from fossil fuels towards renewable energy sources, households will use less gas and oil and shift towards heat pumps and electric vehicles. The fossil fuel use of a household will likely decrease by half, while their electricity use is likely to increase by three or even four. The peak loads that need to be covered by the electricity grids and production, will therefore be much higher and have an impact on the operation and cost of the electricity system.

Energy suppliers and grid operators face some substantial challenges. To avoid heavy infrastructural investments, Linear investigated how these peak loads and its effects could be limited.

Portfolio management

Through Portfolio Management the energy supplier can include the accumulating energy prices in its tariffs. This way tariff control is enabled. The test families were offered 6 tariff blocks per day, which prices daily fluctuated according to the sun, wind, demand and response expected. Each day at 4 PM, the tariffs for the next day were communicated. Next to tariff control, automatic demand response which enables suppliers to switch on household appliances to relieve the grid load, was also investigated.

Wind balancing

The profit gained with wind energy can be substantial. Predicting wind energy can be complex though, due to local differences. These wind predictions are, however, essential to balance the grid. If the actual wind generated strongly differs from the predicted generation, the difference or shortage needs to be compensated. The possibility of automatic demand response, where appliances are switched on or off in real time, was investigated. This way energy suppliers can avoid fines for imbalances.

Transformer ageing

Transformers providing power for districts, have difficulty to locally process peaks of renewable energy. Linear wanted to limit peak loads by better aligning the energy use with the locally generated energy. The project investigated the effect of automatically turning on smart appliances in times of a low primary use.

By downsizing peak loads, loads were spread over time and the temperature of the transformer was reduced. This way either the lifetime of transformer could be improved or a complex aggravation of the transformer was avoided, either way resulting in a cost reduction.

Line voltage profiles

Most household appliances work at a voltage of 230 Volt. When the voltage is low, some appliances will no longer work. If the voltage is high, for example when lightning strikes, some will break down. Grid operators make sure the provided voltage will never pass the allowed limits. The voltage is traditionally highest at the transformer side. The further a building is located from the transformer, the more the voltage decreases due to the fact that the other houses will also use power and due to the resistance of the cable itself.

Since houses are now also generating energy, for example by means of solar panels, the voltage level can now also rise in the middle of a street. For the grid operator this complicates keeping the voltage levels within its limits. This is also the reason why photovoltaic transformers automatically shut down at higher voltages.

We investigated the possibility of automatic demand response to optimise the voltage level at the output of individual buildings. This can be done by specifically switching on the domestic appliances. In doing so we can avoid extra cables to be installed between the buildings and transformer. An alternative without assurance of continuous voltage quality is of course never an option.


EnergyVille's contribution

As the project coordinator, EnergyVille minimalised the effort for the partners and created an open architecture allowing system suppliers and aggregators to partake. We carried out the communication via the website, newsletters and events, we recruted participants, and we were responsible for technical support and system monitoring. Using the Home Lab and Matrix lab, EnergyVille asserted the daily adjustments of the implementation and testing.

As not all required technology was available, we worked out several concepts to prototypes, resulting in the installation of 20 smart boilers and 75 whitewash controllers at the test-household's residence over the duration of 18 months.

Moreover, EnergyVille created simulation environments to measure the impact of the new concepts. Several of these concepts were eventually converted into software used for real-time control of the field test.

EnergyVille's closing responsibility was the data-analysis and redaction of the results.

Wim Cardinaels


Wim Cardinaels

Project Manager Energy Technology at EnergyVille/VITO