University of Twente Student Theses


On the realization of a smart grid demo-site at Coteq in Almelo

Hoekstra, K. (2018) On the realization of a smart grid demo-site at Coteq in Almelo.

[img] PDF
Abstract:The ongoing energy transition in the Dutch electricity grid may lead to grid capacity problems and mismatches between the production and consumption of electric energy. Renewable energy sources such as photovoltaic panels are introduced in residential areas and cannot be controlled. On the other hand, the increasing amount of electric vehicles and heat pumps increase the amount of energy that is consumed significantly. These new devices also provide a certain amount of flexibility to shift their consumption in time. By applying clever control, these devices can tackle the aforementioned problems. This is called decentralized energy management. One of the stakeholders interested in controlling devices in a smart grid is Coteq, a distribution service operator in the Netherlands. To stimulate the development of smart grids, Coteq provides a demo-site containing solar panels, charging stations for electric vehicles and a stationary battery. This research focuses on the development of a smart grid for this demo-site. The simulation and demonstration platform DEMKit is used as a basis to realize a smart grid demo-site. Within this platform we implemented and simulated the valley-filling algorithm, to determine the charging profiles of the electric vehicles. The valley-filling algorithm is split up into an offline part, which determines the fill-level using historic energy consumption profiles, and an online part, which determines the charging power of one or more electric vehicles using the determined fill-level and the current energy consumption of the demo-site. To enable the transition from a simulated environment to a real smart grid implementation, we extended DEMKit to acquire real-time input data for the algorithm. Measurement data is acquired from Influx databases in real time, and smart charging APIs are used to retrieve information about the electric vehicles, such as the state of charge of their batteries and when the charging has to be finished. We tested the smart grid control in a pilot within the demo-site. The added features in DEMKit enabled a straightforward transition from a simulated environment to the pilot. We conducted an experiment where the developed algorithm is used to control the charging of multiple electric vehicles. The output data of the algorithm is communicated to the charging stations to control the charging profiles of the electric vehicles. The experiment resulted in reducing the imported energy by 15% and using 28% more of the energy produced locally instead of feeding it to the grid, when compared to a situation where no smart grid control is applied. Moreover, the peak power consumption is reduced by 23% and the peak power that is exported is reduced by 14%. From the experiment, we conclude that using the valley-filling algorithm to control the flexibility of electric vehicles in a smart grid demo-site results in a reduction of stress on grid assets. The locally produced energy is utilized better and thus less energy is imported through the grid. Also, both the import and export peaks in the grid are reduced compared to a situation without control.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology
Programme:Embedded Systems MSc (60331)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page