VPP implementations: different types of services developed, experiences and platforms

Aggregation by Virtual Power Plants (VPP’s) to provide flexibility to distribution and transmission networks is seen as an important element in the transition to Net-Zero. This paper presents work carried out in the SIES 2022 ERA-Net project, which is investigating in detail the possible provision of flexibility by different technologies. Thus, presented work will be based on real use cases.

One of the partners in the SIES 2022 ERA-Net consortium is the Engineering Technical Centre in Central Scotland (ETC), which has been set up to deliver a technology demonstrator system to manage energy pools using VPP software as well as to investigate how this VPP could operate using a variety of assets in a realistic setting. ETC has interests in two energy pools which are available for immediate deployment in the project:

•ETC's own premises and the wider Scottish Enterprise Technology Park energy infrastructure
•A test area at the Myres Hill wind turbine site

The sites include both electrical and thermal loads, battery storage that can be used for flexibility as well as other consumers in the area.

The VPP design includes the use of cloud-based third-party software for communication, hardware interfaces and an additional pilot VPP software platform providing enhanced services such as optimization scheduling and forecasting amongst other things. Devices from many manufacturers have been incorporated into the demonstrator plant.

The integration of these components and development of the VPP software has proceeded on a learning-by-doing approach. The paper will discuss the design, development of the VPP platform (hardware and software) including a review of the various other platforms that could have been be used in the pilot, In particular, the paper will discuss the challenges with the implementation of the various components and present results on the performance of the design in the context of supplying flexibility services to a TSO/DSO. The VPP software has been running for over a year now and has allowed us to investigate issues with its operation such as reliability, forecasting (Machine learning), optimization (Stochastic and Deterministic) and its potential reuse and design for other projects. Lessons learned and how such a design could be adapted to other potential types of users will also be discussed.