Characterisation of Virtual Power Plants

UoM administered thesis: Phd

  • Authors:
  • Guy Newman

Abstract

The growing number of micro generation devices in the electrical network is leadingmany to consider that these devices can no longer be considered as fit and forget, butshould instead be considered as having a demonstrable network impact which shouldbe predicted and utilised. One of the techniques for considering the impacts of thesedevices is the Virtual Power Plant (VPP).The VPP is the aggregation of all the Distributed Generation (DG) connected into thenetwork up to and including the connection voltage of the VPP, such that thecumulative power up the voltage levels can be seen in the single VPP unit, rather thanacross a broad spread of devices.One of the crucial tasks in characterising the VPP, developed in this work, is theability to correctly predict and then aggregate the behaviour of several technologytypes which are weather driven, as a large proportion of DG is weather driven. Of thisweather driven DG, some can only typically be dispatched with modification and therest cannot be dispatched at all. The aggregation of the VPP as part of the electricalnetwork is also developed, as the constraints of the network and the reliability of thenetwork cannot be overlooked when considering the aggregation of the VPP.From a distribution network operator's (DNO) perspective, these characterisationmodels can be used to highlight problems in the network introduced by the addition ofDG, but are also generally utilitarian in their role of predicting the power output (ornegative load) found throughout the network due to DG. For a commercial agentinterested in selling energy, these models allow for accurate predictions of energy tobe determined for the trading period.A VPP agent would also be adversely affected by line failure in the network, leadingto the development of an N-1 analysis based upon reliability rates of the network,which is used as the basis for a discussion on the impacts of single line failure and themitigation available through feedback from the DNO.

Details

Original languageEnglish
Awarding Institution
Supervisors/Advisors
Award date1 Aug 2011