Demand side response is a method whereby financial incentives are used to encourage customers to lower electricity use at peak times. This enables load, voltage, thermal, balance and other constraints on the electricity network to be managed. In this project we aim to develop methods for responsive control of electric appliances at meso-level. This includes mid-range entities such as small and medium enterprises, communities, campuses, car-parks or similar.
Based on already available data from monitoring equipment previously installed and data to be obtained through planned monitoring, we intend to create a curated database of disaggregated electricity profiles for the different services (such as heating, lighting, electric vehicle chargers, office use, etc.).

We will use data from the three case studies to develop our solution: a car park, a plastic factory and a village community. Based on these profiles, and on different optimisation scenarios, we aim to solve optimisation problems with constraints. This will allow us to overcome issues currently encountered with energy usage inefficiency on meso-level. For example, the use of simple demand response control in a car-park during winter, on a low voltage network which operates at the limits of its capacity, leaves no spare capacity for charging electric vehicles. This happens because heating, lighting and office appliances are on. As a result, vehicles may not receive any charge despite spending the whole day connected to a charger in a car park.

While large scale solutions are developed by large companies such as General Electrics and individual load energy management is an active area of innovation, savings at meso-level are mostly neglected. This is partly because individual monitoring all of appliances would be too costly. Therefore, as the first step towards a solution for this problem, we plan to create a database of typical profiles with confidence bounds at optimal time-scales and granularity. Developing responsive algorithms and practical and cost-effective control systems for our three case-studies will serve as a proof of principle and open more avenues for the use of mathematical and data analytics techniques to create solutions for efficient energy use management at meso-level.

Potential Impact:
The first beneficiary of this research will be the project lead, AND Technology Research (ANDtr). The created service will help increase sales of monitoring equipment and should result in a new, more sustainable source of income through the service developed, which will generate revenue and profit via a monthly subscription model. As a result, ANDtr will be diversifying their offering and accessing new markets.

The secondary beneficiaries are current and future ANDtr clients, for example, car parks management companies, independent factories, and other meso-level organisations and communities. We expect them to benefit in following ways:

- economic benefit will be immediate through optimized energy use, as the service will enable them to avoid peak consumption, which comes with costly penalties. As a result, this will deliver cost saving via smaller energy bills;

- in the longer term, we also expect environmental benefits given reduced carbon emissions through more efficient use of energy.

If the research outputs consistently show, over three case-studies, the success of the methods in energy savings, load balancing or reducing peaks, this could be used to influence energy policy (via DECC) to strongly recommend monitoring and control across different sectors.

As the system will deliver a guaranteed minimal charging level, while keeping designated services such as lighting and heating safe, this project will have direct consequences for user satisfaction. As a result, a wider social impact is expected for example, through improving the desirability of electric vehicle (EV) ownership based on positive experiences. In addition, ANDtr and UoR would facilitate dissemination of the project and results in journals, scientific and industrial communities. This will help to further raise awareness of the benefits, and together may encourage other car users to switch to EVs, leading to larger environmental benefits.

In addition, enabling peak smoothing and better load management (for example in car parks given a predicted steep growth in electric vehicles in near future, and in other settings) will also help implicitly to reduce costs and secure supply for electricity distribution companies.

Finally, in the case of a successful generalisation of this idea to different settings such as school or hospital campuses, small factories and smart buildings, similar economic, social and environmental benefits can be expected on a larger scale.