Virtual Monitoring Data (VM-Data)

Objective(s)

The project will fulfil two key objectives:

Validation and enhancement of the model developed in last year’s LCT Detection NIA project; and
Development of a set of domestic half hourly consumption profiles which can be aggregated and used for virtual network monitoring at feeder level, as well as enabling enhanced network planning and demand prediction.

Problem(s)

The operation of the electricity distribution network is complex and evolving – particularly with growing smart technologies and embedded, renewable generation. The increasing number of ‘invisible’ changes (Electric Vehicles, Embedded Generation, Smart Home, Storage) challenges existing network practices to the extent that the status quo is no longer possible. At present, technology change is outpacing changes in modelling and forecasting of consumer uptake of ‘smart’, Distributed Energy Resources (DER) or Electric Vehicle (EV) technology; therefore, it is difficult to monitor or understand the change in requirements on the LV network, under existing arrangements, without monitoring EV and DER impacts directly at source (or substation level).

Method(s)

The overall project method is to use IBM’s cutting-edge Artificial Intelligence (AI) and cognitive analytics capability to further develop a model developed in the previous “LCT Detection” NIA project, which analyses changes in consumption patterns linked to EV/DER proliferation or other factors. The model will use MPAN-level consumption data from the Energy Market Data Hub (EMDH) plus detailed consumption data (half-hourly intervals or less). The detailed consumption data will also be used to create and refine a set of half hourly customer profiles, which will be used to extrapolate EMDH consumption data into virtual daily consumption profiles, which will be aggregated to achieve virtual feeder and transformer half hourly loading profiles.

• Scope

  The project will establish a data platform and carry out an analytics programme based on data provided by us and ElectraLink. The project will be carried out in three phases:

  Discovery: A business analysis phase incorporating design thinking activities and setting out the work plan for subsequent phases.
  Execution: Five one month long “sprints” of data analytics activity covering two workstreams – Advanced LCT Detection and Profile Modelling.
  Consolidation: Bringing the two workstreams together to produce a final report and model which will enable us to predict load patterns on the LV network.

• Success Criteria

  The success of the VM-DATA project is predicated on the following:

  1. An improved version of the LCT Detection model that identifies Low Carbon Technology (LCT) on the LV network.
  2. Validation results showing, with a high level of confidence, the level of accuracy of the improved LCT Detection model. This will include validation against half-hourly feeder-level readings from our existing LV substation monitoring.
  3. A set of customer profiles, which will enable the extrapolation of ElectraLink’s monthly consumption data to be extrapolated to half hour load profiles on each LV feeder.
  4. Validation results showing, with a high level of confidence, the level of accuracy of the half-hourly feeder loads predicted by the VM data model.
  5. Delivery of an LCT detection and Virtual Monitoring approach that can be transferred into the business as a BAU approach.
  6. Incorporation of the Virtual Monitoring data outputs into one of our tools or network management systems. The project will provide load data at feeder level, with improved accuracy, in a format that is compatible with our systems. The data can be integrated into our existing capabilities/applications which are currently being used by planners/control engineers.

• Potential for New Learning

  The potential for new learning is significant. The VM Data project will develop an enhanced analytics capability; a new operational practice that is directly related to the operation of the electricity system; it will have the capability to support existing and developing industry processes, including dynamic network control, network asset management and DNO installation support for smart installations, thus delivering value for money for customers, and accelerating the development of the low carbon energy sector. The project will disseminate its findings across the GB energy networks sector and beyond. Once proven through the VM Data project, the data analytics and virtual monitoring capabilities will be replicable across all GB DNOs.

• Documents and links

  NIA Project Registration and PEA Document

  Six Monthly Progress Report: Nov 19 – Mar 20

  NIA Project VM Data Closedown Report

• Monthly updates

  • {{update.title}}

  Show more