As electricity is transported across the distribution network some of it is lost. The volume of lost electricity is apportioned across network users according to their responsibility for those losses. The current losses apportionment model, ‘newLAF’, was created in 1995 when distribution networks were almost entirely demand based systems. As more generation connects to the distribution networks, regions such as SHEPD are becoming increasingly generation dominant with periods of net export to the transmission network. There is an emerging risk that the current model was not designed for this background. This project proposes to develop a new losses apportionment model and supporting engineering models using best available data to ensure the apportionment of losses across network users remains accurate and fair.
Benefits
The project is not expected to provide any overall financial benefits. It will ensure that losses on the distribution network are equitably apportioned across all network users.
Learnings
Outcomes
The Engineering models work package determined that the engineering models that are currently used are suitable for use going forward and give an accurate indication of power flows on the network for use in a losses apportionment methodology.
A report has been created discussing these and giving examples of approaches that could increase the accuracy of the modelling in further work in this area. However, work was already underway to improve the calculation of losses on the network so this was not pursued by the project.
For the losses apportionment methodology, a new approach has been taken. The existing ‘newLAF’ model was built for a distribution system where predominantly, power entered from the Transmission system and was consumed by demand customers, leading to a very top-down approach to losses apportionment.
For the new model, both upward and downward power flows were considered. This allows generation driven losses to be accounted for as well as demand driven losses. It means that the methodology is appliable for networks that have demand dominant Grid Supply Points (GSPs) as in the ‘newLAF’ model and, also, networks that have generation dominant GSPs.
Additional engineering input was required to help model how losses for more unusual or legacy network arrangements should be treated. This led to the introduction of some rules that may be SSEN specific and will need to be considered by anyone wishing to adapt this for their own purposes. An example is where the transformation of power is from 132kV to 33kV in most instances, however, it is 132kV to 11kV on occasion. This affects the modelling of power flows when looking at the system as a whole.
The testing of the completed model caused some issues. The initial expectation had been that the results from the SHEPD region would be different from the newLAF model given this network has significant export of generation onto higher voltage levels, but that the SEPD region would show similar results to newLAF given it is a demand dominant network that does not display any export onto higher voltages. This was not the case.
This required careful reassessment of the model to understand if it was correct and why. Ultimately it was determined that the model was correct, but because it handled the data in a different way, it was going to produce different results.
In the end we were satisfied that the model was correct and that we understood how it handled the data.
The new model was presented to Elexon for use in last year’s losses apportionment submission, it was approved and was used to develop the submission.
As a result this project was deemed to have met its objectives.
Lessons Learnt
This project was looking to take a long-standing process and update it for a modern electricity system. Because of the length of time newLAF had been in place, a lot of the detail about the functioning of the apportionment methodology had been lost.
Therefore, when looking at improving existing processes, it is worth ensuring that there is a detailed understanding of the process that is being replaced so that the differences and similarities can be properly evaluated.
