Electric vehicles (EVs) are arguably the biggest game changer impacting the energy sector currently, with their high energy requirements, the need for a whole new infrastructure of charging stations and the challenge of minimising charging peaks.
Technologies such as smart charging and vehicle-to-grid are emerging to address some of these. But for utilities to better manage these issues, greater visibility on their grid is key – as indeed it is for the other challenges arising in a decentralised energy system.
“We are seeing a significant increase in the penetration of EVs and this needs to be managed,” says Carlos Pina Teixeira, CEO and Co-founder of Portuguese startup Eneida in a presentation at the first Engerati Meet on ‘EVs in the grid’. “The impact isn’t clear, but we could be looking at a doubling of peak power demand with impacts such as grid overload, voltage constraints and planning uncertainties.”
LV grid visibility
The first step in addressing EV challenges is to establish the locations of vehicles when they are charging. Eneida’s solution is named DeepGrid and comprises smart sensors which are located on the low voltage feeders and an app-based Internet of Things (IoT) platform.
Teixeira explains that the solution detects the presence of EVs and then determines how many are being charged at any given time using load disaggregation machine learning algorithms.
“These types of algorithms provide a more detailed breakdown of the energy usage at the distribution level. With this information, we can not only assess the impact of EVs on the network but also synthesise it and recommend actions such as flexibility.”
Apps that have been developed to date address use cases including quality of service, energy efficiency, investment optimisation, network modelling and flexibility management.
“The apps can be chosen and uploaded remotely according to the distribution company’s needs,” Teixeira says. “The aim was to make the solution low cost and simple to use.”
Looking ahead, he envisages incorporating other data such as SCADA and smart metering, and the implementation of final customer apps such as on-board diagnostics.
“We want to have a better idea of the impact of EVs but also predict what types of impacts they could have so we can anticipate any surprises.”
EVs as a grid resource
A key use case for EVs is their use as a flexible, dynamic load for utility operators to use to balance supply and demand on the grid. For example, UK system operator National Grid has estimated in its latest future scenarios that by 2040 there could be 34-36m EVs on the roads requiring about 60TWh of electricity per year. However, using EVs as a storage medium could limit the increase in electricity peak demand to as little as 8GW.
In another session at the Meet, Vincent Schachter, Senior Vice President, Energy Services at eMotorWerks, presented data from California showing that most long duration charging takes place at homes and out of a typical 11-12-hour overnight plug-in, less than two hours of charging time is needed for the energy for the average daily commute (50-65km).
In addition, EVs offer the greatest potential for flexibility in the home, exceeding that of any appliances.
“We estimate that about 90% of refuelling is flexible, which creates many opportunities,” he says, noting that while the figures might vary somewhat in other locations, they are order of magnitude.
“The challenge is to choose the intervals used for the charging during those 12 hours and choosing that is a way to optimise the economics and emissions for the EV driver.”
Similarly, there are opportunities during the day with drivers typically leaving the EVs plugged in during their work hours, although currently only about 10% of the long duration charging is occurring at the workplace.
IoT platforms for EV charging
eMotorWerks headquartered in California and acquired by Enel in 2017, has developed a smart charging solution comprising JuiceBox charging stations and the JuiceNet internet of things platform for aggregation of EV loads and the optimisation of charging and provision of services to the grid. These solutions are applicable to residential and workplace as well as public infrastructure applications.
Schachter says that California was a favourable environment to launch the solution, as it comprises about half the EVs in the US – about 370,000 to date, due to strong incentives – and a system operator which has been progressive in allowing distributed resources to participate in various markets, including the real-time and day ahead markets, while frequency regulation provision is in pilot.
At this time several thousands of EV drivers across the state are signed up for smart charging programmes, either directly with eMotorWerks or through partner utilities or aggregators.
Schachter comments that a key feature of the solution is that the driver interfaces should be simple and easy to use, and that level of interaction should be minimal but always allowing an override to default settings such as the usual time the vehicle is required for use in the morning.
“We want drivers to have a nice, easy charging experience,” he says.
Incentives are also proving important, both to encourage EV adoption as well as smart charging participation. For example, in the partnership with the local electricity provider Sonoma Clean Power, up to $4,000 per household is available in subsidies towards the EV and a free smart charger is available to all customers. Almost 2,000 chargers have been delivered with an 82% smart charging opt-in.
Schachter says that the learnings from California are just the start and through the Enel acquisition these are being extended into Europe. In addition, it has enabled the solution to be advanced by integrating Enel’s products such as bidirectional chargers.
“Ultimately it's not just about EVs. With the growth of other distributed resources in the home, if we can co-optimise them together it will create more value to customers and to the grid itself.”