EVs and the grid
Can our grid cope with the extra charging load if we all switch to electric vehicles? A recent report by Australian company Evenergi examines the issues.
If well planned, the transition to electric vehicles will lead to a significant reduction in greenhouse gas emissions, along with other potential benefits such as improved GDP and lower costs for consumers (in the longer term). But can the grid cope? According to a recent report, the short answer is yes, and there can even be positives for utilisation of grid assets and grid stability. But some planning is clearly needed to ensure we capture the benefits without significant issues.
The report, funded by ARENA, the SA government, SA Power Networks and Evenergi, considered both the capacity challenges and renewable energy opportunities for EVs on the South Australian grid.
The report’s main focus is on charging ‘hotspots’. It considers the charging scenarios that might lead to capacity issues (such as a concentration of ultra-fast chargers, for example) and then identifies approaches to avoid or mitigate the issues. Of course, renewable energy plays a big part in this.
A charging hotspot is a place where a significant charging load may occur that is beyond the capacity of the network, whether that’s the limit of the street transformer, the feeder or a substation. The risk of a hotspot occurring depends on many things, including charger type (higher capacity chargers will have a greater impact), charging timing/length, driving distances of EV owners (which affects the amount of charging they need to do) and the location and density of chargers in an area. The report looks at the potential impacts of residential, workplace, fleet and public charging.
The report evaluated the risk of charging hotspots in different locations, and the potential to mitigate that with local solar generation or demand response. For example, it found that there’s a mid-level risk of charging hotspots occurring at the residential level, but a high chance of using rooftop solar or demand response to reduce the risk. As another example, the report identifies a low risk of hotspots at shopping centres as they are likely to use lower power chargers, with the shopping centre owners unlikely to upgrade them to higher power chargers if that requires significant connection upgrades/cost. Highway chargers are identified as having a high risk of becoming a hotspot—new network connections will be sized appropriately, but multiple charging stations on a long feeder could result in capacity issues.
In one interesting example that shows the result is not always predictable, London’s Waterloo bus garage thought it would need a network upgrade to cope with 51 electric buses, but, after the network upgrade got delayed, it found it could work within the limits of the existing connection by topping up buses during the day and only fully charging off-peak.
Managing capacity issues
While there is a potential for hotspots, there are many ways to reduce or eliminate the risk without needing significant network upgrades. There are three main strategies: price signals, infrastructure solutions and smart charging.
– Price signals
A price signal encourages charging at times when there’s less likely to be a capacity impact. Price signals can be static or dynamic. Static signals, like time-of-use pricing, have proved effective in the USA, Germany, UK and Japan. Evenergi reports
their own experience in the UK, where EV drivers actively seek out low-cost charging overnight. However, the report warns that they are a blunt instrument—they can be too effective, ineffective or have unintended consequences. For example, they can lead to too many vehicles charging at the same time, with peaks shifting to off-peak times (see Figure 1). Controlled load meters with lower tariffs, as currently used for hot water, are another option, and have the flexibility of being controlled by the network provider. In the past, such loads have been set to operate overnight at off-peak times, but the timing could be randomised or set to the daytime when there’s excess solar PV available. Dynamic prices or rewards that vary per day or per customer can also help randomise the load. Dynamic pricing trials are currently underway in Norway and California, with the latter trial finding success via an approach where prices are posted one day ahead.
– Infrastructure solutions
In a recent ‘thoughtpiece’, the UK’s grid operator National Grid proposed to build a network of super-fast 350 kW chargers located at similar sites to current petrol stations. If EVs could handle this charging rate (most currently can’t), these chargers could provide about 100 km range in four minutes. National Grid projected that just 7000 of these stations with seven to nine chargers each would be sufficient—slightly less than the number of petrol pumps currently in the UK—and that this would reduce the need for private charging at home and work. The result would be less need to upgrade home connections and local networks; it would also provide charging capability for drivers without off-street parking.
The advantage of such an approach is that the charging load can be planned for and located to manage capacity issues.
Another infrastructure solution is pairing large-scale battery storage with charging stations. The report suggests this could be a way to deliver grid services and synchronise with over-generation of renewables to provide a positive impact on the wider grid.
Vehicle-to-grid is another possible approach, with signals used by the power network to charge/discharge vehicle batteries according to the needs of the grid. The Nissan Leaf is enabled for vehicle-to-grid and trials have been conducted; the report notes that current combined capacity of Leafs would be over 200 MW globally.
Another option, of course, is to upgrade the network to make it better able to cope with variable loads like EV charging. The report discusses a UK Power Networks project called Active Response, which enables sharing of power between feeders and substations which it expects will “materialise 3.5 GW of peak demand and save customers £271 m by 2030.”
– Smart charging
Finally, and perhaps most promisingly, there’s smart charging, where the utility controls the timing of the EV charging. In its simplest form, this can be without knowledge of the vehicle’s state of charge or the owner’s driving intentions. Alternatively, it can be ‘paired control’, which allows a user to guarantee their car’s charge level and opt out of charge throttling events if the car is below a certain charge level. A pilot in Toronto, Canada, has illustrated the effectiveness of this approach (see Figure 2).
Whether you’re talking about vehicle owners charging at off-peak tariffs or grid operators throttling charging, control over the charging process is needed. The controller for this can be in the car—many EVs have charge controllers that enable them to avoid peak pricing events. More sophisticated controllers can even accept dynamic signals from the market. Alternatively, the controller can be in the wall charging device, the EVSE. These can simply manage charging times or accept dynamic signals to modulate charging, not just on-off control.
Signals can be based on time, peak load or grid frequency, or can be dynamic from the market operator for load shedding events. There can be value in this for vehicle owners—they may be able to avoid high demand tariffs or get paid for putting power into the grid. This can be particularly valuable for fleet operators.
Coordination with renewables
The report highlights that SA is the first region on the national electricity market that has so much rooftop solar that minimum demand shifted from overnight to midday; AEMO has forecast that SA may have negative minimum demand by 2025–26 when you factor in wind and large-scale solar. This means under-utilisation of capacity at certain times of the day (excess solar in the middle of the day and excess wind at night). Electric vehicles could play a significant role in improving utilisation and hence stabilising the grid by shifting their load to these times of minimum demand.
At the residential level, the benefits of charging from rooftop solar with and without a battery are also explored in the report, via analysis of several homes and their current load profiles; as might be expected, larger solar systems can reduce demand. Charging stations can also be beneficially paired with local solar generation, if there’s suitable roof space or if a large solar canopy can be constructed.