With the increase in renewables and demand - why is scheduling important?
In our latest blog, Mark Burkley, Head of IoT at Lightsource Labs discusses how electric vehicles are an ideal complement to renewable energy given the intermittency of renewables however the detached usage and consumption in electric vehicles highlights the need for a HEMS system to manage the scheduling.
The electrification of road vehicles is well underway and sales of Electric Vehicles (EVs) are growing rapidly. Much has been written about how the grid infrastructure isn't ready for widespread EV charging, but it could also just as easily be said that EV charging isn't ready for the grid. EV chargers need to be made aware of the optimal time to charge to make the best use of renewable energy, to use energy when it is the lowest price, and to avoid putting excessive demand on the grid. A simple timed schedule will soon be no longer viable and the use of an Energy Management System (EMS) to coordinate charging intelligently will be needed.
Supporting the charging infrastructure presents challenges for national grid operators who are also having to deal with the increased amount of renewable energy produced for the grid. Renewable energy is intermittent by its nature - the wind doesn't always blow and the sun doesn't always shine. This presents challenges for grid operators who need to maintain backup infrastructure to be ready to meet demands when renewable energy sources are not available. Both of these challenges can actually complement each other however and have the potential to work together to create a stable and sustainable grid that supports large-scale EV charging. EV charging is a unique electrical load in that the time at which energy is consumed from the grid differs from the time the energy is actually needed due to the battery storage in an EV. This makes EV charging an ideal complement to the intermittency of renewable energy that can partially solve the potential misalignment of renewable energy production and grid energy demands. Managing charging schedules is not a trivial undertaking however and highlights the need for an EMS to automate the scheduling.
Optimising EV charging manually is a complex task as there are many variables. Dynamic energy tariffs, where different rates apply at different times during the day, weather forecasts and predicted consumption all needs to be considered. An EMS can automate automatically optimise the charging by analysing dynamic energy prices, factoring in weather forecasts and using Machine Learning (ML) to learn about typical usage patterns of an EV and other connected loads.
Annual electrical power production in the UK in 2020 was 313TWh [1] of which 43% was from renewable sources. The average load factor of the UK grid was 72%. Interestingly, losses due to transmission and distribution in 2019 were 25.4TWh [2] or about 8% of the energy produced. This loss value should drop as more "prosumers" come online and consume power locally instead of importing from the grid. Capacity from all producers was 77.9 GW in 2019 [3]. Of this, renewable capacity continues to increase and reached 22 GW in 2019, despite being de-rated to account for intermittency. Scheduling loads to occur during periods of high renewable energy production and low demand would increase the
ratio of renewable energy and lower the de-rating figure used for renewable sources.
As of 2020, there are 38 million licensed vehicles in the UK [4]. Of these, 432,000 or a little over 1% are classed as Ultra Low Emission Vehicles (ULEVs). Of these, Battery Electric Vehicles (BEVs) accounted for 64% of new ULEV registrations in 2020 which is an increase of 169% over 2019. The trend toward electrification of road vehicles is clear, but this is not without its challenges. If we imagine that every vehicle on the road today was electric and every one of them were to begin charging at the same time at a rate of 7kW, then the grid would need to supply 114 GW which is beyond the capacity of the national grid. On the other hand, if we consider that there were 356 billion miles driven by road in the UK in 2019 [5] and if we take an estimated EV consumption value of 240Wh/mile then the grid would need to supply 87 TWh per year to keep them all charged. So while the peak power demand may seem impossible to supply, the total energy required is not as significant. If the demand for charging power was spread out then it would only require 10 GW which is quite possible to accommodate, and is in fact less than the current capacity for wind power alone. From this we can see the blocking issue to the electrification of every vehicle on the road is not capacity, but scheduling.
The theoretical figure of 87 TWh would appear to require an extra 28% to be added to the grid capacity. The grid is currently only 72% loaded which, purely in theory, means the capacity is already there and no extra capacity is needed to electrify every vehicle on the road. Further, the derating currently applied to wind and solar production due to intermittency could be reduced by consuming power exactly when it is available. In reality, more capacity is still needed as the grid can't operate at 100% efficiency all the time. But the addition of new capacity is happening anyway, particularly with the addition of more renewable energy sources and improved utilisation of existing production
capacity will improve the load factor.
To date, less than 1% of the UK vehicle fleet are fully electric. While this number will undoubtedly rise rapidly, particularly as the sale of new diesel and petrol is proposed to be discontinued from 2030, grid balancing of EV charging is not an immediate issue. But optimising charging to occur at periods of high capacity and low pricing is still very beneficial to EV owners. If EV owners are also prosumers with solar panels or other local generation capability then factoring in solar
irradiance forecasts is a further optimisation. Applying AI and ML via and EMS to predict how much charge is required for the day ahead further optimises the cost of charging.
