The US Energy Information Administration (EIA) suggests that the multi-year historical average for transmission and distribution losses in the US power grid are approximately 5% annually (average of annual losses in 2014 through 2018).
More recent data for 2018 suggests that the loss has increased to 6.6%, which, by any account,is significant. Put in perspective, the actual loss in 2018 amounted to 275 TWh. The average retail price of electricity in 2018 was $105.30/MWh, meaning that this loss was valued at $29 billion. The increase in loss from about 5% to more than 6.6% has resulted in increased costs of loss by almost $7 billion.
What is causing the increased loss, and what, if anything, can be done to manage it?
Loss is caused by passing a current through a conductor that has resistance. The resistance converts some of the power that is passed through it into heat, just like an electric toaster or heater. But in the case of the grid, the temperature increase in the conductor is small. The temperature rise at any point in the conductor may be small, but the total energy loss is large over distance and time. The bad news about loss is that it increases with the square of the current passed through the conductor, so a system that is delivering 100 watts, with a loss of 5 watts, would increase the loss by a factor of four if the power delivered was doubled. In the case above, increasing power to 200 watts would increase loss to 20 watts—four times the loss at 100 watts and double the previous loss at percent loss at 10%.
The US electric grid delivers power to meet the demand of customers—demand that varies dramatically every day. The grid is designed to be capable of meeting the annual peak demand with redundancy, but the average capacity over an entire year is about half the annual peak. The average loss is based on the overall use and hence is focused approximately on the average use.
What this means is that the loss at peak demand is four times higher than the average loss. The following three cases, performed on a software simulator, demonstrate the impact of demand on loss. At 50% capacity, there is a 5.6% loss for a 5 MW load. Increasing the power delivered to 10 MW reveals an increase in loss from 0.28 to 1.1 MW—almost four times the loss at 50% capacity.
These two losses are calculated as a fraction of the total power delivered. However, more significant is the marginal loss that occurs for a small increase in power delivered.
Increasing the load by a small amount above the 10 MW load reveals that the loss for a small addition in capacity is more than 18%.
If the grid loss averages about 6.6%, peak demand loss might be 13%, and marginal loss may be as high as 25%. If the delivery loss is higher, as is often the case where the storage and return is remote and takes place between utilities, the marginal loss at peak may be even higher.
Clearly, loss increases dramatically with the demand, and at high loads, it may be a significant part of the power that is dispatched.
What does this mean to a power system operator?
- To deliver more energy to customers, we either need to increase the capacity factor on the existing system, or we could build two or three new systems equal to what exists. The most likely selection is to increase capacity factor, resulting in an accompanying increase in losses. The challenge will be to manage and minimize that loss. The most efficient delivery of energy takes place when the capacity is constant and voltage is managed to minimize losses caused by the delivery of reactive power. Enbala has demonstrated a unique capability to optimize a distribution feeder by optimizing grid edge assets for precise control of voltage and reactive power on the line.
- The most efficient way to deliver energy over a long period is to deliver at a constant rate. This delivers maximum energy at minimum loss for the total energy delivered. Any deviations above and below a constant level will marginally increase total percentage losses.
- The use of peaking generation that is remote from loads is subject to high marginal losses when used for peaking. The use of other utilities to store and return energy (as California does with BC Hydro, providing afternoon storage that is returned at peak demand times) is subject to these high losses.
- When done near the load site, the use of grid edge management technology to manage demand or connected DER devices can result in almost no loss at all.
The strategy to minimize loss can be summarized in a very few steps:
- Manage demand at the grid edge to be as near to constant as possible, reducing peak and valley periods through the use of some form of storage.
- Manage voltages to optimize the delivery of power to be provided close to unity power factor. This is a more complex concept that will be addressed in detail in a coming paper.
We’d be happy to talk with you more about any of these thoughts and recommendations—or to answer your questions about how Enbala’s Concerto™ grid balancing platform can help you minimize system loss.