UK Power Networks

This paper presents a novel approach to remote power flow prediction in distribution networks using Machine Learning (ML) models trained on Phasor Measurement Unit (PMU) data. The research addresses the challenge of maintaining Distributed Energy Resource (DER) operation during communication failures with centralized Active Network Management (ANM) systems. We evaluate three advanced machine learning models-LightGBM, Temporal Fusion Transformer, and TSMixer-against traditional baseline approaches using high-resolution PMU data collected over seven weeks. The models were tested for both short-term (1-hour) and medium-term (4-hour) prediction horizons, with the Temporal Fusion Transformer achieving the best accuracy (RMSE of 21.83A and MAPE of 3.89% for 1-hour predictions). However, considering the practical constraints of edge deployment in distribution networks, LightGBM emerges as the most suitable solution, offering comparable performance (RMSE of 22.59A) with significantly lower computational requirements. The results demonstrate that local intelligence powered by machine learning can be used to maintain DER operations during communication failures, potentially reducing unnecessary curtailment and supporting grid stability.

The Active Response project, led by UK Power Networks, aimed to address the challenges posed by the increased electrification of load due to the adoption of electric vehicles (EVs) and other low carbon technologies (LCTs). From 2018 to 2023, the project focused on demonstrating hardware and software solutions for distribution feeder management and optimisation using innovative power electronic devices (PEDs) called Soft Power Bridges (SPBs) and Soft Open Points (SOPs) manufactured by Turbo Power Systems (TPS). These devices were optimised during live operation using a Distributed Energy Resource Management System (DERMS) provided by Smarter Grid Solutions (SGS) and Resource Innovations (RI). The project was completed in two phases, with the first phase focusing on offline optimisation and proof-of-concept, and the second phase on operationalising the trials. The trials demonstrated significant benefits, including improved network capacity, enhanced fault level and voltage control, cost savings, and environmental benefits. Key learnings include the effectiveness of high voltage (HV) and low voltage (LV) switching, the importance of LV monitoring, challenges with SOPs and SPBs, and the integration of live data.

The increasing penetration of inverter-based resources in Great Britain's power system has resulted in a decline in system strength, which has given rise to sub-synchronous oscillations. Traditional methods of assessing system strength are inadequate

As levels of distributed generation and storage continues to grow in distribution networks, there is key focus on increasing hosting capacity – the level of headroom available on distribution networks to connect more low carbon technologies. One aspect of hosting capacity is maximum fault level, which must be constrained below design limits to maintain a safe and secure network. The utilisation of bus-section reactors is one method of creating such headroom. This paper provides background to the technology, and explores challenges identified from system planning, operation, and network analysis. Future recommendations are made to guide the utilisation of schemes for future usage in increasing fault level capacity.

The proliferation of Electric Vehicle (EV) charging, Heat Pumps (HP), and Photo Voltaic (PV) generation within low voltage (LV) distribution networks is expected to escalate. This growth will cause congestion, triggering reinforcement needs to increase hosting capacity (HC) and maintain power quality and continuity. This paper presents findings from a Phase Switch System (PSS) deployment trial on the UK Power Networks (UKPN) distribution network in London and Brighton. The patented PSS increases HC by dynamically balancing three-phase LV distribution mains (LV-main) through real-time phase switching of conductors. With 2023 loads, results from the trial LV-main that supplies 40 customers, increased HC by up to 232A and 202A on average. Also, it provided an estimated 28 MWh annual energy savings in losses reductions. Crucially, simulations of future 7kW electric vehicle charging loads indicate that over 5 years, the PSS could provide 31A per phase more HC than a new and additional cable rated 500A per phase. This comparison to new cables is made, as the HC they provide is widely understood. The trial showcased the PSS as a flexible, rapidly deployable, cost-effective solution to reduce losses and unlock HC, for low-carbon technology integration into LV networks.

Abstract Dispatch of Flexibility Services is defined as the process through which the DSO informs a flexibility provider of the required level of service within operational timescales. This chapter provides more context on how processes for flexibility service dispatch have been developed and co-ordinated.

Abstract Distributed energy resource's inherent flexibility can be a significant enabler in facilitating a faster and lower cost connection to the GB system. This chapter introduces the concept of Flexible Connections and how these have helped facilitate the connection of significant volumes of zero carbon generation in GB. The relationships and interactions of Flexible Connections and Flexibility Services are also discussed.

Heterogeneous networks have emerged as a popular solution for accommodating the growing number of connected devices and increasing traffic demands in cellular networks. While offering broader coverage, higher capacity, and lower latency, the escalating energy consumption poses sustainability challenges. In this paper a novel optimization approach for orthogonal heterogeneous networks is proposed to minimize transmission power while respecting individual users' throughput constraints. The problem is formulated as a mixed integer geometric program, and optimizes at once multiple system variables such as user association, working bandwidth, and base stations transmission powers. Crucially, the proposed approach becomes a convex optimization problem when user-base station associations are provided. Evaluations in multiple realistic scenarios from the production mobile network of a major European operator and based on precise channel gains and throughput requirements from measured data validate the effectiveness of the proposed approach. Overall, our original solution paves the road for greener connectivity by reducing the energy footprint of heterogeneous mobile networks, hence fostering more sustainable communication systems.

This chapter provides a wide analysis and outlook on aspects related to optimally planning and operating multicarrier energy systems under the integrated local energy community (ILEC) concept. It touches upon the main challenges and provides a consolidated framework of existing and recommended practices when designing and planning the multicarrier energy system, taking into consideration uncertainties and operational aspects before taking decisions. At the same time, it covers aspects of optimal operation in shorter timeframes and real-time control, by putting forward the paradigm that ILECs can serve multiple objectives by exchanging services with the wider energy system as well as serving local market incentives. Additionally, this chapter entails potential data architectures required to enable real-time visibility and controllability of multicarrier energy systems. Finally, the chapter underlines the holistic approach when planning and operating the ILECs and how these two activities need to be met for efficient multicarrier energy systems.

This paper presents the ConnectMore tool, developed by EA Technology as part of the Innovation Project, Charge [1], funded by Ofgem's Network Innovation Competition. SP Energy Networks (SPEN), partnered with EA Technology, PTV Group, and Smarter Grid Solutions (SGS) to develop a self-service tool that supported customers seeking to install new EV chargepoints on their electricity distribution network. Combining the specialist domain knowledge of the project partners, with SPEN's network data, the ConnectMore Tool enables optimised geospatial planning of EV charging point locations between now and 2050. This is achieved through combining forecast EV energy requirements with modelled grid capacity, and the cost to implement the proposed new connection. Where a flexible connection is technically viable, the tool will identify this to SPEN, potentially enabling customers to secure a flexible connection more quickly than a firm connection would otherwise require. The tool allows customers to quickly test multiple locations on the distribution network, receive an estimated cost in minutes, rather than weeks, reducing effort and costs for SPEN, and hence reducing overall costs for customers, whilst accelerating the deployment of new charging points and maximising utilisation of the available grid capacity.