Algorithms for Efficient, Resilient, and Economic Operation of Pre-Emptively Reinforced Reconfigurable Distribution Substations

Stochasticity of demand profiles at electricity distribution substations is increasing due to the proliferation of low carbon technologies; in particular mobile, bi-directional, or intermittent loads such as electric vehicles and heat pumps. The decarbonisation of heat and transport will cause a long-term increase in overall connected load, making substation reinforcement necessary, whilst planning of upgrade locations and capacities remains challenging. This project will investigate pre-emptive substation reinforcement with algorithmic topology control, to utilise the additional installed substation capacity only when required.
Distribution Substation Dynamic Reconfiguration (DSDR) proposes the installation of additional transformers in parallel with the existing transformer in each substation, removing the need to scrap and replace these. Telematics-controlled switches are installed on the high- and low-voltage side of each transformer in the substation, with local agent algorithms deployed to control in real-time when each parallel transformer is brought into or taken out of service. Substation reconfiguration is thus controlled to optimise for maximum operating efficiency. The threshold algorithm most recently trialled in medium voltage parallel transformer substations is implemented as a baseline, and a novel model-based reconfiguration algorithm is proposed, implemented, and evaluated in software and hardware.
This work led to a 1.34% improvement in algorithm performance on substation efficiency, over a yearly demand profile including residential and new electric vehicle load for the year 2050, equivalent to a potential saving of 2.68 TWh annually if deployed UK-wide. This approach unlocks several opportunities to operate existing substations in the smart, flexible, resilient, and efficient manner that will be required to reach the net zero target by 2050.