The Anticipating phase of the project builds on the work undertaken in the Understanding work stream.

The  objectives in the trials associated with the anticipation stages of the TVV project will be to utilise the data gathered in the monitoring trials of energy consumption to predict the likely energy usage in the future.

The primary questions being answered by undertaking these trials are:

  • How can improved modelling enhance network operational, planning and investment management systems?
  • To what extent can modelling reduce the need for monitoring and enhance the information provided by monitoring?
  • To what extent can modelling be used in place of full network monitoring? How might modelling assumptions change over time?

Network Modelling Environment

A key component of the TVV project is the creation of the Network Modelling Environment (NME). This system will replicate the functions of an existing Geographical Information System (GIS) which displays the electrical network on a topographical map. However the NME will interface directly with the Distribution Management System to replicate changes made on the network into the DMS without the need for rework. The NME will also act as an asset database.

Furthermore the NME will give operators and planners the ability to run specific scenarios on selected areas of network to model future constraints based on likely scenarios of energy usage.

The initial trials of the NME will verify that the modelling techniques are applicable on real Low-Voltage networks and will then be expanded to incorporate time-varying functional requirements into these scenarios.

Once the basic modelling techniques have been proven, the NME will be used to model the investment, planning and operational requirements for a range of disruptive technology being installed.

These include carbon saving devices and micro-generation systems that many consumers have installed or are likely to install into their homes and businesses in the next 10 years. This could include the wide take up of electric vehicles, additional solar generation, heat pump installations as well as any new energy saving or carbon reducing technologies that come to market.

The final trial will assess if the headroom can be enhanced by changing set points of existing voltage control devices at primary substations, without violating constraints at other, related parts of the network.

Distribution Management System

The TVV project has set up a Shadow Distribution Management System (DMS) to monitor and control technology being deployed onto the Low-Voltage network in Bracknell.

All monitoring equipment uses this Supervisory Control and Data Acquisition (SCADA) system and future electrical equipment installed onto the network will be sent control signals from this system.

The first trial of the system will investigate the optimal level of control in the distribution management system as opposed to the SCADA systems used for higher voltage networks. At present there is very limited monitoring of the Low-Voltage network and no control functions are associated with the management of these networks.

The second trial in this phase of the TVV project will focus on an assessment into the readiness for business as usual considering the points below:

  • The DMS can act as the coordination hub for network management which will integrate with various intelligent distributed energy resources to be deployed in the LV network, leading with demand resource and battery storage resources.
  • The ability to calculate where and when additional resources can be used to re-enforce the current network during peak demand times.
  • The ability to charge LV storage units during off-peak times in order to make them available during peak times without impacting current demand.
  • The availability of power analysis information based on load profiles for estimation of current system demand.
  • The availability of power analysis information based on substation monitoring information available from the monitoring solution detailed previously.
  • The ability to link into the Honeywell ADR as an aggregator of demand response across an estate of buildings to create a despatchable demand resource.
  • Systematic evaluation of telecommunications solutions in NTVV and other available Smart Grid projects.
  • Assess the DMS solution as a functional stepping stone for any distributed device on the network [such as voltage control schemes and power flow management], enabling outputs to flow into DNO and third party systems.
  • Evaluate technical, economic and commercial performance of tool in releasing capacity on the network.

Aggregation 

We are looking to understand the blend of customers down a feeder, whether that is different domestic classes, different business types, or indeed a mixture of the two, to more accurately predict network demand. Experiments with real smart meter data show
that small groups of households (up to 200) are not able to aggregate in such a way as to cancel out and smooth out (mollify) peaks in demand. Depending on the modelling outcomes based on the characteristics of the properties on the Low Voltage (LV) networks, the aggregation will fall into one of three categories:

• Red: Including volatility the demand on the network at peak is expected to be close to or exceeding capacity and needs addressing immediately or near future
• Amber: Demand is expected to be moderately high at peak and is volatile enough to require monitoring, or will reach such a state in the near future

• Green: Demand is expected to be low and consistent (low volatility) such that monitoring is not required.

One of the project aims is to understand what characteristics and drivers cause a LV  network to be green, amber or red under various scenarios (for example, roll out of new technology or changes in demand). Clearly there could be substantial cost savings by
avoiding monitoring all LV networks. But, perhaps more importantly, this method provides a much better understanding of consumers than substation monitoring alone and consequently a much smarter grid.

Forecasting

The TVV project will use real network data to generate forecasts and run scenarios against the network in Bracknell. These forecasts will be used to validate against the actual network performance . we will apply consumer-focussed mathematical forecasting techniques to improve distribution network planning. The project will also create an agent-based forecasting model, enabling short, medium and long-term demand predictions with the confidence factors in a RAG status as described above. The TVV also aims to forecast the impact of potentially disruptive technologies or load intensive events on network behaviour and the associated network investment requirements

The Mathematics behind the TVV project

The TVV project has been working closely with project partners at the University of Oxford to utilise advanced mathematical techniques.

These techniques will inform how to best utilise the data that distribution network operators will obtain from smart meters and other grid connected telemetry devices.

The work on characterisation, forecasting and buddying could revolutionise the way that network operators manage and maintain the electricity networks of the future.

It is anticipated that mathematics will play a role in devising efficient, effective and scalable methods and insights which could underpin the next generation smart LV grid technologies.

This video below gives an overview of these techniques and how they are being applied in the project