Another successful event! The 4th LV workshop and the Thames Valley Chamber of Commerce energy seminar were held at the university of Reading on the 1st and 2nd of March. This two-in-one event aimed to share recent developments in demand forecasting and energy storage control, and bring the academics and industry together for discussions and collaboration. Below is a summary of the presentations given during the 4th LV workshop and the TVCC seminar.
4th LV workshop
In the continuation of the theme of the previous workshops and close links with the New Thames Valley Vision (NTVV) project, the 4th LV workshop aimed at the academics working on the applied research on demand forecasting techniques and energy storage control in the Low Voltage (LV) networks.
The first day was dedicated to forecasting techniques for power systems, starting with demand forecasts at the national level and then focusing on demand at the LV feeder and house level.
Dr Jeremy Caplin (National Grid) gave an informative and fascinating presentation on the state-of-art forecasting of the electricity demand at the national level, including an insight in to how weather and major events impact both, embedded renewable generation and demand.
Dr Stephen Haben (University of Oxford) and Dr Laura Hattam (University of reading) presented on the forecasting techniques for demand and loading from Low Carbon Technologies at the LV level developed for the New Thames Valley Vision (NTVV) project. Stephen’s talk highlighted the complexity of forecasting at the household and the feeder level driven by the range of customer behaviours and the temporal range of the forecast. Laura’s talk focused on forecasting the uptake of Low Carbon Technologies (EVs and PVs) as clusters and modelling the uncertainty of their output.
Dr Stuart Galloway (University of Strathclyde) presented on the forecasting of demand for single buildings on LV network, focusing on the analysis of the errors and highlighting the inter- and intra-day correlation of errors. Abstract:
“Electrical loads such as those in islanded power systems, on residential feeders and in large individual buildings require increasingly descriptive load profile representations in order to capitalize on operational innovations including dynamic tariffs, embedded storage and demand response programs. Accordingly, deployment of Smart Meters and new innovations in Building Energy Management Systems deliver measurements to inform models for anticipating these. Although usage time and its relationship with load magnitude at both inter and intraday timescales can support forecasting applications and the general representation of load, variability at this scale presents several modelling challenges. Conventional day ahead load forecasting models for small and building level loads are utilised coupled with a boosting procedure based entirely on the dependency structure of the intra-day error. Models are devised to be operational with only a small training set allowing near immediate use. Several non-Gaussian modelling strategies to accommodate the complex intra-day forecast error dependency structure are demonstrated on operational data taken from residential LV networks and several non-residential buildings, showing improved forecast errors over a variety of short term time horizons.”
Dr Florian Ziel (University Duisburg-Essen) presented and compared several short-term forecasting techniques for LV feeders, focusing on how feeder size impacts the accuracy of the forecasts and the sensitivity of forecasts to temperature.
Dr Kavin Narasimhan (University of Surrey) presented how concepts from social practice theory have been applied to agent-based modelling of energy demand at a household level. The concepts are based on three components of social practice (meaning, skill and material), linking the activities with energy usage and validating how the relationships between activities and energy usage change the over years.
Dr Richard Snape (De Montfort University) introduced the SWELL project and presented findings from the demand response trials based on control of electrical heating in response to demand and generation forecasts. The findings highlighted the effectiveness of the chosen demand response business model on benefits for the customers and how customers responded to the trial. Abstract:
“Recent work has highlighted the need for smarter and more localised management of electricity networks, with the UK National Infrastructure smart power report recommending that “Enabling the transition to more actively managed local networks should be a government priority. By Spring 2017 DECC and Ofgem should consult and set out how and under what timeframe this transition should best take place”
This talk presents an approach to managing local renewable generation and load, particularly legacy electrical heating (storage heater) load, in a community energy scheme. It reports on a recently completed case study project which has installed equipment in 50 buildings to implement such a scheme and record data about usage and the success of local balancing, the community business model and smart controls in combination with resident interaction. Finally, the talk describes a model incorporating learnings from the project which may be further used to investigate the replicability of this scheme in communities across the UK.”
The second day of the workshop focused on energy storage from two perspectives: energy storage for network support and commercial applications of energy storage ( as part of the TVCC energy seminar).
The morning session aimed at the energy storage for the network support. Nigel Bessant (Scottish and Southern Electricity Networks) have set the scene for the network support session by presenting DNO’s perspective on the energy storage and gave a brief overview of existing innovation projects involving energy storage.
Dr Timur Yunusov (University of Reading), Gideon Evans and Josh Martin (Scottish and Southern Electricity Networks) presented on the development and control of battery energy storage within the NTVV project. Gideon and Josh gave an overview of the project, focusing on the specification requirements for the Energy Storage and Management Units (ESMUs), learning points from the installation and commissioning of the units at the street level, and from the integration of control system into the existing IT environment. Timur focused on the automated control system for the ESMUs, providing the overview of the control methods and results from the trials of control methods on ESMUs installed on LV feeders.
Andreas Procopiou (University of Manchester) presented an overview of the challenges in PV rich LV networks and how actively managing residential PV and battery energy storage systems impacts the operation of PV-rich LV networks. Abstract:
“European countries have promoted through different incentives the use of residential-scale photovoltaic (PV) systems resulting in a significant number of installations in LV networks. However, since LV networks are traditionally designed to cope with demand without controllable elements, with higher PV penetration levels, technical problems such as voltage rise or thermal overload of assets are expected to occur. Indeed, distribution network operators (DNOs) are now required to either proceed with costly network reinforcements (e.g., upgrading transformers, conductors) or transit towards the active management of LV networks which can be an attractive and cost-effective alternative.
In this context, the industrial project “Active Management of LV Networks” funded by Électricité de France R&D and run by The University of Manchester was set to investigate the adoption of active network management techniques to tackle voltage and congestion issues. This talk will present the key findings from two techniques (1) the coordinated control of residential-scale PV systems adopting already embedded power control functions (e.g., Volt-Var, Volt-Watt) and (2) the control of residential-scale battery energy storage systems in a way that provides benefits to both customers (i.e., self-consumption) and DNO (i.e., voltage management). These innovative solutions that require limited network information and observability have the potential to increase the ability of LV networks to host residential-scale PV systems without the need of expensive and time consuming network reinforcements, paving the way for a cost-effective low carbon future.”
Dr David Greenwood (Newcastle University) presented on the recent developments in control techniques for LV networks from the Newcastle University, including results from field trials and laboratory experiments. Abstract:
“In electrical networks, Energy Storage can fulfil multiple functions across a range of voltage levels. The research in this presentation focusses on simulation and control of energy storage in low voltage distribution networks, enabling power flow management, voltage control, and integration of domestic photovoltaic systems. We investigate coordinated control of multiple energy storage systems, and of energy storage in conjunction with demand response. We describe methods for mass simulation of LV networks, and heuristic planning techniques to investigate the location and sizing of storage systems. Our results are validated through energy storage demonstration, carried out as part of the Customer Led Network Revolution project.”
Thames Valley Chambers of Commerce Energy Seminar
The TVCC seminar had a great combination of speakers, representing industry, academia and investors. Audience from the 4th LV workshop was joined by members of the TVCC and the representatives from industry interested in commercial applications of energy storage.
The first presentation was by Dr Ben Potter (University of Reading) on energy storage research highlights. Ben described recent developments in energy storage technologies and costs, followed by two examples of energy storage applications (reducing network connection costs and frequency response). Ben also highlighted that other key factors (i.e. control methods, security and forecasting) play an important role in ensuring that the application of energy storage is successful.
Nigel Bessant (Scottish and Southern Electricity Networks) presented on the role of energy storage from DNO’s perspective, described the importance of energy storage performance and gave a couple of examples how energy storage could perform a key role in DNO to DSO transition.
Dr Matthew Rowe (DNV GL) gave an overview of energy storage on the global stage, highlighting the idea that energy storage could be seen as a disruptive technology with a potential to provide a wide range of applications.
Dr Marek Kubik (AES Corporation) presented on the frequency response application of energy storage at the Kilroot power station, which proven to be more effective than conventional plants in delivering frequency support.
Neal Kalita (Energy Locker) gave an insight from developer’s perspective into the investment process for energy storage, highlighting the key factors to be take into the account and the key stages in the process of decision making for the energy storage investment.