Every year 130.000 people meet on a field near Roskilde to experience the festivities of Roskilde Festival. The sudden influx in people takes its toll on the electrical grid, as the facilities which provide food, music and more use a lot of power. In fact, during peak hours Roskilde Festival are forced to run diesel generators in order to keep up with the overall power consumption.
In order to reduce the need for these diesel generators and eventually get rid of them entirely, our project is looking to shave peak power consumption through load management. Specifically, we are looking to regulate a cooling container. The idea is simply to cool down the cooling container to a lower temperature than what is required, for example -30C instead -18C, thereby storing the energy thermally. We can then turn off the cooling containers and decrease the load on the grid when power is most needed..
In order to achieve this, we’ve written an algorithm to control the cooling container. The algorithm looks at data from previous days in order to predict when the power peaks, this is needed in order to cool down the cooling container in advance. Power-meters set up on the festival then feed the controller with live-data, and when the power exceeds a certain threshold, the algorithm tells the cooling container to shut off, until the temperature inside the cooling container once again reaches -18C.
The model estimates that peakshaivng with cooling container could eliminate 3 hours of power usage during peak time, this would save around 3kWh according to our model. Also the most efficient way to peak shave is simply to use all the, over night, accumulated thermal energy immediately once we are in peak time (usually from around 12:00).