Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - Spatialec (Developing methods to model local area temporal domestic electricity demand)

Teaser

Driven by the need to de-carbonise energy supplies, to reduce overall demand, to shift demand away from systemic or critical peaks and to cope with localised, time-specific and/or intermittent generation, the emerging view of Europe’s (green) smart grid future places...

Summary

Driven by the need to de-carbonise energy supplies, to reduce overall demand, to shift demand away from systemic or critical peaks and to cope with localised, time-specific and/or intermittent generation, the emerging view of Europe’s (green) smart grid future places substantial emphasis on demand-side response (DR) as a key component of a sustainable electricity system.

As many stakeholders have noted, there is a clear need to develop a modeling framework which can allow regulators, policy researchers and commercial analysts to understand how particular combinations of scenarios might affect electricity infrastructures at varying levels of geographical scale and at different times of the day, week and year.

Previous work in this area has tended to rely on ‘average customer type’ or ‘average power’ based approaches which prevent the analysis of the full heterogeneity of potential demand response behavior for different customer groups. This is especially relevant for the dimensioning of local power networks if some types of customers are much more (or less) likely to show substantial response than others and if these customers are also concentrated in particular areas.

This research responds to this need by developing a microsimulation based approach to demand modeling at the household level and integrates this with known ‘spatialisation’ approaches to develop area level temporal demand profiles. The envisaged outcome is therefore a model that can produce not only a local area temporal demand (and potential response) ‘map’ under a range of scenarios but also indicators of the range of responses likely as input to network operator infrastructure management analysis and/or local power generation planning decisions.

This work will apply the University of Otago’s ‘Energy Cultures’ and ‘GREEN Grid’ research to develop local (neighbourhood) level electricity demand models. The results will be validated using a range of data sources and methods before being used to analyse the local demand implications of a range of NZ demand response scenarios such as EV uptake, energy efficient lighting or increased heat pump use.

In the return phase the modeling approach will be applied to similar data derived from the four year (2014-2019) ‘SAVE’ Low Carbon Network Fund project to produce a generally applicable model for the South East of England. This model will be validated using area level substation monitoring data before being used to model the local area effects of a number of demand response interventions being trialed by SAVE. The work will be disseminated widely with a view to ensuring exploitation through relevant infrastructure and demand response stakeholders and also through the development of a future research programme with selected European partners.

Work performed

The work has implemented a national and local level electricity demand model for New Zealand. Main results achieved so far are:
- the archival of a clean version of the GREENGrid household electricity demand monitoring data with the UK Data Archive (https://dx.doi.org/10.5255/UKDA-SN-853334 - (Anderson et al. 2018)) for third party re-use. This data archive is supported by an R package prepared by the Researcher (https://cfsotago.github.io/GREENGridData/ - (Anderson and Eyers 2018));
- analysis of the contribution of lighting, hot water, space heating (via heat pumps) to demand profiles;
- analysis demonstrating that the literature on energy demand reduction is beset with under-powered studies using biased samples that can rarely be generalised. The work (Anderson et al. 2020) uses the GREENGrid dataset to provide guidance to enable analysts, managers and decision-makers to significantly improve the evidence base;
- development of a complex reweighting approach to provide national level estimates constrained by NZ Census 2013 population totals. This work was re-used in a report for the New Zealand Energy Efficiency and Conservation Authority (Anderson 2019b) to demonstrate the value of the approach in analysing scenarios of future peak demand;
- the development of this model into an ‘instantaneous hot water and heat’ model that could be used to explore future electricity demand scenarios in the context of New Zealand’s renewed policy focus on a low emissions economy and thus switching most space heating and light vehicle mobility to electrical power;
- an analysis of the potential consequences of the uptake of LEDs, of curtailing peak demand and of switching to more efficient heating (heat pumps);
- an analysis of the charging patterns of a sample of up to 50 domestic electric vehicles in partnership with a NZ EV owners club (FlipTheFleet). This work has produced both academic outputs (Ben Anderson et al. 2020) and contributed to a substantial report to the New Zealand Parliamentary Commissioner for the Environment (Myall et al. 2019);
- development of an Area Unit level temporal electricity demand model for the regions of Taranaki and Hawke\'s Bay. These results formed the basis for a report prepared for the New Zealand Energy Efficiency and Conservation Authority (Anderson 2019b) and for an end-of outgoing phase results workshop. They have also contributed to a conference paper on the future role of electric vehicles in balancing the New Zealand gird (Ben Anderson et al. 2020).

Final results

Progress beyond the state of the art includes:
- analysis of the components of residential peak electricity demand in New Zealand. This novel analysis using GREENGrid household electricity demand monitoring data provided the NZ Energy Efficiency and Conservation Authority with insights and though leadership on potential new policy foci;
- the archival of a unique multi-circuit household electricity monitoring data set via the UK Data Archive\'s ReShare service. This dataset has already seen substantial re-use.
- the development of open source tools to enable the modelling of local area temporal electricity demand scenarios. These have attracted substantial interest from electricity distribution businesses who currently have little access to smart meter data and so find modelling peak demand for capacity planning extremely challenging;
- the development of practical guidance for energy research managers and analysts on how to design effective energy intervention studies. The work has already had a significant impact on the design of the future Building Research NZ (BRANZ) national household energy efficiency study (HEEP2).

Expected results to the end of the project include:
- the application of the tools and methods developed in New Zealand to the UK context. This will involve the application of the methods to existing \'smart meter like\' datasets held at the University of Southampton. These models will use evidence derived from the literature, from the NZ GREEN Grid project and from the SAVE project trials;
- the development of a programme of exploitation/impact based on the results of the work. This will involve further engagement with NZ and UK/EU electricity/energy distribution businesses to understand how the results, insights and tools developed during the work can be best mainstreamed into business as usual;
- the development of an international network of applied researchers concerned with the modelling of local (neighbourhood) level demand scenarios building on the NZ-focused network that formed around the end of year 2 results workshop held at the Otago Energy Research Centre Symposium 2019.

Website & more info

More info: http://www.energy.soton.ac.uk/tag/spatialec/.