Power outages due to extreme weather, wildfires and other factors have created a great deal of interest and publicity concerning the reliability and resilience of the electric grid. These are important topics that will play a changing and valuable role as the grid transitions to meet these challenges and evolving customer energy needs.
There are several ways in which major failures occur. I will deal with more of the causes in the next few blog posts and will discuss how distributed energy resources (DERs) can mitigate or prevent failures and support the grid in times of need.
There have been multiple large outages that have occurred in the last 50 years. Initially, it was believed that most were a result of a loss of generation capacity and underfrequency resulting from the loss. Other threats have been identified. Perhaps the most unusual was an outage in Quebec that was caused by a large solar storm. This may be a growing threat, particularly as the geographic areas covered by our interconnected grids grow in size.
Much of U.S.-based generation relies on steam turbine technology, including coal and nuclear-powered generation. Steam turbine machines are huge — and presumably robust. The steam turbine system — invented by Charles Parsons in 1884 — is based on the original principle but have been improved and optimized to transfer maximum power from steam to the generator.
Steam is injected into turbine blades at a high temperature and pressure. The blades are positioned to maximize the power transfer to the mechanical system at the designed speed of operation (50 or 60 Hz). If the rotating speed of the generator is either increased or decreased, the angle at which the steam strikes these turbine blades changes. People that fly airplanes know that if the aircraft is slowed, the angle at which the airflow strikes the wing changes, and at a point, the flow over the wing changes abruptly from a smooth flow to a turbulent flow. This is known to pilots as a stall. The steam turbine experiences similar issues, and if the frequency falls below a certain point, the flow of steam through the turbine blades becomes turbulent, and damaging vibration can occur. The generator is automatically tripped offline, and a shutdown sequence is initiated.
In a large grid, if there is a large loss of supply capacity, (loss of transmission or generation), the frequency may decline rapidly. When that happens, if the decline is not halted quickly, steam turbines connected to the grid will trip below a specific frequency in order to protect the turbines. Once started, the process may cascade to other generators. Restarting may take an extended period, as these machines must be stopped, cooled, and a restart initiated. In some cases, this may take several days. In a nuclear generating station, there may be additional time needed to bring the generation capacity back to its operating target.
I remember asking a nuclear plant operator what happened to the reactor steam when the generator suddenly shut down. His response: it is vented outside. It makes a huge noise, and if it is winter, no one can go home, as cars in the employee lot will be quickly covered with a thick layer of ice.
A cascading failure can occur quickly, leading to a total collapse. This is generally far too fast for any manual intervention. It may start in one location and depending on frequency settings at other steam turbine plants, subsequent generator trips may occur many miles away. This is a serious issue, and there are systems in place to halt any rapid drop in system frequency.
When a loss of capacity occurs, the frequency will begin to decline. The initial decline is slowed by system inertia. The utility term used for the Rate of Change of Frequency is RoCoF. Interesting to note that the highest inertia among generators are the steam turbines, as they rotate at the highest speed (60 Hz – typically 1,800 or 3,600 RPM). Hydro plants, with huge heavy rotating mass (up to 1,000 tons) have the lowest inertia constant, as they generally rotate slowly (<500 RPM). The inertia constant is proportional to the square of the speed of rotation. Solar and wind generation have almost no inertia component. As steam turbines are displaced with renewables and gas turbines, system inertia will decline, and this may create serious issues for system operators
One can examine the frequency response to a loss of generation and can understand the composition of the grid. The graph shows three responses to a loss of generation. The blue line is the Eastern Interconnection, which has a large component of steam turbine generators (high system inertia). The black trace is WECC – the western grid, which has both steam and hydro generation (medium system inertia), while the green is ERCOT (Texas), which has many gas turbine generators and a large component of renewable capacity (low system inertia). System inertia plays a key role in the frequency response after a loss of capacity.
Utilities have installed systems to address any sudden frequency decline. Many utility substations are equipped with a protection system that will monitor frequency and will trip entire distribution feeders if the frequency falls rapidly. These systems may operate in utilities that are far from the loss of capacity. Widespread use of systems of this type have demonstrated that many severe disturbances can be handled by the grid, without leading to a cascading collapse.
The National Electricity Reliability Corporation (NERC) has established standards to be met by all interconnected utilities in Canada and the U.S. NERC requires the system to be capable of maintaining stability with the largest probable single loss on the system. This standard has successfully reduced failures.
Distributed energy resources are a great resource for preventing system collapse because of their ability to achieve rapid load reductions, either by deploying battery capacity, or by shedding demand. Some utilities pay well for this support. A large battery was installed in Australia, claiming that it would pay for itself in a short time. Skeptical as I was, I watched, and it indeed did pay for itself faster than expected, based NOT on trading energy as I had assumed, but on providing fast Frequency Control Ancillary Services (FCAS).
Services of this type will become more important with time. As coal-fired generation declines, replaced at least partially with renewables, system inertia will decline. But as long as there are steam turbine generators on the grid, the need to respond quickly will be valuable. The ability of “behind the meter” systems to respond autonomously at times when rapid response capability is critical will provide real value to both the owner and the grid operator.