Dr Euan Morris, Research Associate, Electronic and Electrical Engineering, University of Strathclyde
22nd November 2021
A common definition for a smart local energy system (SLES) is one that is ‘capable of managing a wide variety of complex interactions between end users, physical devices which make up the network, and the data flows which connect them’.
But for a truly ‘smart’ SLES, the design must be for more than a one-off or bespoke solution only applicable to the circumstances it was created for to provide value for money and to support scale-up. The design needs to fulfil the following three criteria:
- Flexibility – through its lifetime, be being capable of dealing with evolving use-cases, connection to additional and new devices (whilst losing some along the way) and expansion to include other energy vectors such as transport or heating.
- Scalability – be expandable to facilitate both increased demand and types of demand, as well as new sources of generation and types of generation.
- Reusability – a SLES solution designed for one problem should be applicable to other similar problems without the need for a new bespoke solution to provide value for money.
In short, there is a strong need for ‘plug and play’ capabilities for SLES to truly capture their potential. While technological limitations are often spoken of as a barrier to SLES, there are existing, mature, AI technologies which allow for these ‘plug and play’ capabilities to be added to existing systems.
To demonstrate the capabilities of these AI technologies, researchers at the University of Strathclyde, Coventry University, and the University of Sheffield have worked to extend an existing smart grid demonstrator ADEPT including enabling a range of small SLES units to buy and sell electricity between each other rather than using the national grid. Details on other aspects of the ADEPT extension work are discussed here: EnergyREV researchers extend the ADEPT Smart Local Energy System demonstrator.
The results of this case study have demonstrated that the AI techniques applied enabled cost savings for all market participants compared to ‘normal' operation without this trading capacity.
The case study has demonstrated the need for the three ‘smart’ SLES aims presented above:
- Flexibility was demonstrated as additional energy trading functionalities were added to an existing system.
- Scalability was shown because the deployment and testing of multiple units was possible without the whole system requiring reconfiguration.
- Reusability came from the implementation, as our design would be applicable to a wide range of units wishing to take part in energy trading.
Using similar approaches can bring us closer to true ‘plug and play’ implementations and ensure full capabilities of SLES can be captured.
The full case study is available here: A plug and play artificial intelligent architecture for smart local energy systems integration