Network Integration of Renewable Energy
Background
The UK has a range of obligations that require an increase in the penetration of renewable energy generation. While the 2010 targets are quite modest (10% in England and Wales, 18% in Scotland) targets for later years are significant: the Scottish Executive is proposing a target of 40% by 2020.
The location of renewable resources and the size of projects means that many schemes will be connected to distribution networks. As these networks were not designed to accept distributed generation (DG), their connection creates a wide range of technical problems. While a range of options exist to mitigate adverse impacts, the costs can be substantial making potential schemes less attractive and, in some instances, impeding development of renewables.
Edinburgh's research programme on network integration is aiming to assist the achievement of renewable generation targets by exploring the challenges that distributed generation brings and developing appropriate solutions. The early programme, a collaboration with UMIST (now University of Manchester) was supported by EPSRC under the Renewable and New Energy Technology (RNET) for three years until September 2003. The programme has progressed significantly and forms a key part of the EPSRC Supergen V Network Infrastructure consortium.
Network Impacts
Historically, distribution networks were designed to supply customers from the national grid and were operated passively to ensure customer's electrical supply quality. Connection of distributed generation fundamentally alters the operation of distribution networks. The changes and impacts are well-documented and include those shown in Figure 1, below.
These impacts are examined when the developer applies to connect with Distribution Network Operators (DNOs) assuming worst-case operating conditions to ensure no reduction in quality of supply. Typically, worst-case conditions occur with the generator operating at full capacity whilst local load is at a minimum: these create the greatest local voltage rise which tends to be the most significant factor.
Several techniques may reduce adverse network impacts but these are project specific and depend on the issue: voltage rise is currently addressed through network and generator operational changes or through asset upgrades. These measures allow DG connection but may be expensive in generator revenue or capital cost terms. The added capital cost can adversely affect project economics with charging methods compelling developers to finance the expenditure upfront or face enhanced use-of-system charges. Another major issue is that under the current first-come first-served policy for DG development it is possible for a quite minor project to prevent development of larger sites, 'sterilising' parts of the network. This risk will only increase as the volume of development grows. Our programme of work has focussed on these two key issues.
Evaluating Network Capacity
A key method of minimising these problems is for DNOs to issue guidance to developers regarding the existence, or otherwise, of spare connection capacity. To do this, DNOs need to ascertain the capacity of new generation that may be connected to their distribution networks. Our approach work has been to use Optimal Power Flow (OPF) to maximise capacity at specified locations. Initial work using proprietary software modelled DG as negative loads to capture DG operation at fixed power factors with the capacity of the network evaluated by maximising capacity through load addition. The operation and application of this 'reverse load-ability' technique is explained in detail in [1]. Recent work has incorporated fault level restrictions within the OPF technique [2] allowing use on urban meshed networks and the eventual aim is for a tool that can readily assess available capacity subject to all relevant technical standards.
The development of bespoke routines has allowed the OPF to be altered to allow explicit modelling of generator control schemes. The incorporation of intelligent generator operating strategies has allowed initial assessment of the impact of widespread local voltage control [3].
The OPF techniques allow the network’s limiting factors to be highlighted (e.g. equipment thermal ratings) and further work in Supergen V is aimed at providing an efficient and effective means of determining network upgrades and reinforcement that allow further DG to be accommodated [4].
Key Publications
Several papers relating to network integration of renewables is available from my publications page. Key work includes:
[1]
G. P. Harrison and A. R. Wallace, 'OPF evaluation of distribution network capacity for the connection of distributed generation', IEE Proc. Generation, Transmission & Distribution, 152 (1), January 2005, pp. 115-122. View the draft.
[2]
P. N. Vovos, G. P. Harrison, A. R. Wallace and J. W. Bialek, 'Optimal Power Flow as a tool for fault level constrained network capacity analysis', IEEE Trans. Power Systems, in press. View the draft.
[3]
P. N. Vovos, A. E. Kiprakis, G. P. Harrison and J. R. Barrie, 'Enhancement of Network Capacity by Widespread Intelligent Generator Control', Proceedings 18th International Conference and Exhibition on Electricity Distribution CIRED 2005, 6-9 June 2005, Turin, Italy. View preprint.
[4]
P. Vovos, J. W. Bialek and G. P. Harrison , 'Optimal generation capacity allocation and network expansion signalling using OPF', Proc. 39th Universities Power Engineering Conference, Bristol, Sept 2004. View abstract.
|
