EV clustered charging can be problematic for electrical utilities
September 4, 2017
September 4, 2017
The electric vehicle (EV) market is booming. Bloomberg estimates EVs will account for 35% of new vehicle sales worldwide by 2040. The increase in electric cars means the load on electricity systems are also on the rise. This poses a problem for electric utilities around the world. New electric vehicles such as the Tesla Model X and S batteries are capable of storing approximately 100-kilowatt hours (kWh). Compare that to the electricity usage of an average home is approximately 25-kilowatt hours (kWh) per day. This means the electricity grid will need to support an increase to four times the current load.
For this reason, the Rocky Mountain Institute has recommended that utility companies prepare for the rapid adoption of EVs to avoid undue stress on the distribution network. By anticipating and planning for the growing load of charging EVs, the institute says utilities can both accommodate the increased load and gain benefits for the entire system.
Why electric vehicles could pose a problem for electric utilities.
Electric utilities are seeking solutions to problems resulting from clustered electric vehicle charging. Electricity generation is not an obstacle, instead it is at the local level, the so called ‘the last mile’, that the problem occurs.
Local distribution grids are not built to accommodate the huge spikes in demand where electric cars will be particularly prevalent. Transformers, which connect every home and business to the power grid, are the most vulnerable and affected elements of the system. Most residential transformers serve between 10 and 50kVA of load, a single plug-in vehicle (PEV) with a 240V (Level 2) charging system consumes approximately 7kVa.
Add more than one electric car to the same local transformer (charge clustering) and overloading is more likely to occur. This causes damage to the electrical equipment which reduces the lifespan and can result in outages and added costs. The risk of overloading local transformers is particularly high during peak hours. When all electric vehicles owners in a single area recharge at the same time.
Studies suggest that higher penetration of EVs can increase transformers’ loss-of-life factor, by up to 10,000 times. As an example, the Sacramento Municipality Utility District has recognized 17% of their transformers need replacing as a result of EV-related overloads, at an average estimated cost of $7,400 per transformer.
As the EV market grows, the opportunity for demand-side management solutions will continue to rise. The net effect of demand response is to ease system constraints and generate security and economic benefits for the market. The need to manage EV load is greater than ever, and the question utilities are facing is where to start.
Evaluating solutions to local electrical overloading resulting from EV charging.
The world’s first smart-charging pilot program to use vehicle-side data on a range of models had encouraging findings. The pilot program was successful in demonstrating how paired smart charging can meet the needs of EV owners and those of the electric utility to protect the network.
Many factors will influence EV load management and load profiles including:
- Commute cycles,
- Proportion of city versus highway driving
- Use of renewable energy sources
- Available rate structures
- Incentives for home charging
- Deployment/availability of public infrastructure
- Types of EVs popular in the utility’s service territory
To know which solution is appropriate for a utility’s EV load management strategies, utilities must gather data from its own customers. Data collection may take many different forms, but using local data to develop a strategy has merit. The utility customer is not generic.The ChargeTO pilot program included six months of vehicle monitoring followed by five months of paired smart-charging. The initial monitoring period focused on capturing the default charging
The ChargeTO pilot program included six months of vehicle monitoring followed by five months of paired smart-charging. The initial monitoring period focused on capturing the default charging behavior of EV owners. The paired smart-charging period evaluated the ability of the system to reduce or defer EV charging load. The goal was to understand the technical and non-technical requirements for a successful and scalable smart-charging program.
Findings reveal problematic EV charging behavior is common.
Baseline data revealed three peak load periods during weekdays corresponding with EV owner charging. The first peak load period was due to EV charging which occurred immediately upon returning home. The second peak load occurred shortly after 7 pm to coincide with an off-peak time of the time of use (TOU). A third peak load occurred at approximately 6 am from ‘delayed-departure’ which programs EV charging to conclude before a required departure time. The resulting peaks were at approximately 5:30 pm, 8:00 pm, and 6:00 am.
In addition to peak load times, the pilot program identified six distinct EV owner types. These are segments of commuting and non-commuting vehicle use for each of the three EV drivetrain types, including plug-in hybrid EV (PHEV), short-range battery EV (srBEV), and long-range battery EV (lrBEV).
Electric vehicle clustered charging could cause problems for electrical utilities.
The EV owner segments revealed what would happen if more than one EV of similar owner types were to charge on the same local transformer. In these scenarios, the transformers were rated for 100 kVA and the utility was comfortable with a 30kW incremental load to account for a typical household load.
Local distribution transformers receive ratings for average loads over a 24 hour period. Transformers are capable of handling higher than average loads for short durations. Repeated, prolonged, or significant overloading will lead to a shorter life of the transformer, triggering the internal fuse which causes a local outage, or in the worst case a complete failure.
An analysis of clustered charging for short-range battery electric vehicles demonstrated overloading in some situations. Clustering of long-range battery electric vehicles proved to be a significant risk of overloading.
The study revealed a cause for concern for clustering of even non-commuting long range BEVs. The data pointed to days of the week where a load of 45 kW, which was problematic even for this larger 100 kVA distribution transformer. If this were a 50 kVA distribution transformer this type of load would be even more so problematic.
There is a significant cause for concern when looking at clustered charging of commuting long-range BEVs. In this scenario, every day of the week charging load was over 70 kW, getting as high as 95 kW on one weekday. This presents a critical issue as infrastructure will fail or its service life will be significantly shortened.
EV paired smart-charging mitigates the risk of local grid overloading.
The second half of the paired smart-charging program demonstrated the ability to reduce peak charging loads by half. Also, 85% of EV charging load could be ‘shifted’ up to two hours (instantaneous curtailment capacity). In addition, the pilot project discovered that 70 to 80% of the charging load could be shed at peak load times in case of an emergency and still ensure that all vehicles were fully charged in time for a scheduled trip. The ability to reduce and shift peak load could be the key to mitigating the risk of costly and dangerous local grid failures.
The case for electric vehicle smart charging.
The challenge of supporting EV charging is not isolated. Electric utilities worldwide are looking for methods of distributing charging events across the full span of the off-peak period. Managed charging, also known as smart or intelligent charging, entails a combination of infrastructure and communication signals sent to a vehicle or via a charger to influence the driver’s decision on when to charge the car. Currently, the technology is in the state of early deployment, already 69% of utilities in the United States are considering implementing programs.
Staying ahead of electric vehicle charge event load patterns.
Electric utilities will need a complete understanding of the shape and size of EV charging load on their network to manage the problem. With monitoring, utilities can forecast when and where to upgrade infrastructure, and shape charging load with price signals or direct control.
Both load management and pricing approaches to shaping demand are growing in popularity as an alternative to simply building larger infrastructure. Collecting and analyzing local data and understanding how EV load profiles will differ due to several factors will provide the information necessary to make the optimal selection of building and load shaping solutions.
SmartCharging pilot program conclusions.
While increased production is manageable for the grid on a large scale, localized problems will arise. Clustered charging, when multiple electric vehicles charge in the same area, is a risk to local infrastructure. Ways of managing local load such as building infrastructure and static price signals will prove to be less effective than before. The dynamic nature of shifting load as a result of vehicle charging is more challenging to predict and manage. Electric utilities need new solutions to mitigate the risk to infrastructure.
Handled correctly, electric vehicles offer new, beneficial possibilities for electric utilities. Dynamic price signals and smart charging technology can help avoid building more infrastructure. The grid of tomorrow can be one which rewards beneficial charging behavior and driver input to manage load distribution. Scalable smart-charging has proven more successful at walking the tight-rope between the needs of utilities and it’s users. This at a fraction of the cost compared to previous methods of managing localized grid load.
ChargeTO provides a useful template that utilities can use to capture their baseline EV load, understand the overloading risks, and evaluate load shaping solutions such as paired smart-charging. The project also proves that paired smart-charging can be a useful tool for supporting widespread EV adoption while also benefiting the grid.
Findings from the ChargeTO pilot program have directly influenced the FleetCarma and Con Edison program, SmartCharge NewYork. A program designed to help EV owners reduce the cost of charging and enhance electric grid efficiency and resiliency, making service more reliable for everyone.
Through FleetCarma, EV owners are able to track EV stats and earn rewards by charging in Con Edison service territory. EV owners also earn rewards by shifting charging to off-peak hours and staying clear of summer peak hours.
FleetCarma supports the adoption of electric vehicles by providing the technology and services required to evaluate the feasibility, plan for the adoption, and support the operation of electric vehicles. Fleetcarma offers solutions which include the SmartCharge Manager, SmartCharge Rewards, and Electric Vehicle Research Platform.
Eric Schmidt is the Marketing Manager of EV Ecosystems at FleetCarma, a division of Geotab. He has over 10 years of experience helping Canadian technology companies tell their stories to the world. His work in marketing, design, communications, public relations, print, video, advertising, data analysis, and research has helped increase the awareness of FleetCarmas unique set of products and services. Prior to becoming interested in business, technology, and new energy Eric graduated with Honors in Graphic Design and Advertising from George Brown College.