UKERC Energy Data Centre: Projects

Projects: Projects for Investigator
UKERC Home >> UKERC Energy Data Centre >> Projects >> Choose Investigator >> All Projects involving >> NIA_SPEN0022
 
Reference Number NIA_SPEN0022
Title Weather Normalised Demand Analytics (WANDA)
Status Completed
Energy Categories OTHER POWER and STORAGE TECHNOLOGIES(Electricity transmission and distribution) 100%;
Research Types Applied Research and Development 100%
Science and Technology Fields PHYSICAL SCIENCES AND MATHEMATICS (Applied Mathematics) 75%;
ENGINEERING AND TECHNOLOGY (Electrical and Electronic Engineering) 25%;
UKERC Cross Cutting Characterisation Not Cross-cutting 100%
Principal Investigator Project Contact
No email address given
SP Energy Networks
Award Type Network Innovation Allowance
Funding Source ENA Smarter Networks
Start Date 01 August 2017
End Date 01 December 2018
Duration 16 months
Total Grant Value £249,000
Industrial Sectors Power
Region Scotland
Programme Network Innovation Allowance
 
Investigators Principal Investigator Project Contact , SP Energy Networks (100.000%)
Web Site http://www.smarternetworks.org/project/NIA_SPEN0022
Objectives Analyse historical demand data and create normalisation models Evaluation of the weather normalised demand model Completion of uncertainty analysis against historical data. Analysis of demand trends - weather related vs customer behaviour. Creation of summary report and data files. Delivery of final presentation or paper at industry event/conference. The project will be considered successful if the aforementioned objectives are realised and the outputs become a useful part of SP Energy Network’s suite of planning tools.
Abstract Forecasts of electrical load are used by network operators to determine the volume, type and location of investments. Load forecasts are based upon the power flows through substations and are adjusted for the embedded generation on the network. Currently, there is no regular adjustment made for the effect of weather upon demand in the local area served by each individual substation. This means that it is extremely difficult to separate out the effect of weather upon demand and the effect of other customer behaviour upon demand (e. g. energy efficiency measures, increases in the number of electric cars being charged, the closure of industrial premises, etc.)This results in additional uncertainty when making investment decisions leading to under or over investment in individual network areas and therefore suboptimal outcomes for customers. Additionally, it leads to inconsistent regulatory reporting. An example would be a mild winter causing demand to drop and the load index metrics becoming artificially low for that year. Weather patterns and customer behaviour are two key drivers in electricity demand. By undertaking this project, it will be possible to better understand how these have changed historically within a given licence area and will provide invaluable insights into future demand scenarios. It will also highlight their relative significance and current trends. This will allow asset managers to develop better and more targeted investment strategies. Furthermore, accurate demand models will provide more realistic data for investment risk and cost-benefit analysis and subsequently lead to better returns for customers. A comprehensive understanding of the effect of weather upon demand is a key enabler for the transition to a DSO model. An advanced numerical weather prediction model will be used to simulate hourly weather conditions. The location-specific weather data will be weighted by population density and aggregated across the areas served by each of SP Energy Network’s primary substations. This will be used to create a weather normalised demand model. An embedded generation module will be implemented within the demand function in order to reduce generation-related uncertainties. The demand model will be trained against recorded data for the period of analysis using advanced machine learning algorithms. A comparison of the demand model results vs historical data will provide a level of confidence in the predictions. The results will then be used to infer weather vs customer-behaviour trends over the period of analysis and predict future demand scenarios.Note : Project Documents may be available via the ENA Smarter Networks Portal using the Website link above
Publications (none)
Final Report (none)
Added to Database 17/12/18