Understanding the Price of Flexible Operations

As more and more variable generation comes online, a new EPRI tool quantifies the costs of operating thermal generation more flexibly

By Chris Warren

It’s hard to overstate how dramatically renewable generation has grown over the past decade. According to the International Energy Agency (IEA), the cumulative global installed renewable electricity capacity in 2012 was a little over 1500 gigawatts—most of it hydropower. By 2021, a combination of policy support, steady cost and technology improvements, and increasingly urgent efforts to address climate change had resulted in record-breaking annual additions to renewable generation capacity.

According to the IEA, 2021 saw nearly 295 gigawatts of new renewable generation, primarily wind and solar. The IEA forecast 2022 would be another record year, with renewable additions reaching 320 gigawatts despite supply chain challenges and inflation that has caused solar and wind prices to rise recently.

While additions of renewable energy are important to address climate change, they also raise questions about the steps necessary to maintain grid reliability. Questions of escalating importance as society looks to electricity to power transportation, heating and cooling, and a growing list of industrial operations. Indeed, as electrification becomes more critical to a decarbonized economy and society, grid reliability becomes increasingly paramount.

Flexible Operations Stress Thermal Power Plants

As variable generation sources increase, one of the most pressing questions is: What are the potential impacts on existing thermal generation assets? Many of these assets, including natural gas- and coal-fired power plants of different sizes and turbine types, are already operating differently than they were designed to do. For example, the influx of variable generation resources and changes to energy markets have forced many baseload fossil fuel plants to cycle frequently and operate in a load-following manner.

These new modes of operation place stress on thermal assets fundamental to maintaining grid reliability. Significantly, they can also elevate the possibility of expensive and disruptive unplanned outages. “As we continue to transition to this clean energy future where we have more variable renewable energy, the demands on thermal dispatchable assets to perform effectively when the sun isn’t shining and the wind stops blowing is very high,” said Stephen Storm, an EPRI technical executive. His research focuses on fleet generation optimization. “We depend on them for frequency controls and to be able to turn on and off reliably and as efficiently as possible.”

But for utilities charged with maintaining grid reliability and prioritizing operations and maintenance (O&M) investments, moving to more flexible operations can be an argument against funding upkeep and maintenance of thermal assets. “It can be very difficult to show that you need increased budget dollars or need to fund projects when you are generating fewer dollars in megawatt sales,” Storm said. “Instead, folks in charge of budgets may say you’re getting 30% less than you did last year because you generated fewer megawatts.”

The Cost and Uncertainty Management Tool

Past revenue and other metrics don’t accurately reflect the risk of equipment stress and damage and unplanned outages from flexible operations. It’s far better to base investment decisions on a quantifiable assessment of the costs of increased cycling. That was the idea behind the development of EPRI’s Cost and Uncertainty Management Tool, which was released late in 2021 and is already being used by utilities worldwide.

The tool is designed to quantify the costs of flexible operation via frequent cycling for eight different gas- and coal-powered plants and also forecast both equivalent forced outage rates and the incremental costs of increased flexible operation. The tool builds on and improves an existing simplified cost model and seeks to expand the traditional understanding of the economic and reliability impacts of plant cycling.

In the past, cycling costs were calculated by considering the price tags of startup fuel, additional power plant staffing hours, and the lower efficiency that comes with the increased heat rate associated with cycling. But the development of this new tool and related EPRI research resulted in a more nuanced view. For example, more cycling reduces the lifespan of components, and their failure can lead to more frequent forced outages. “All of these starts and stops can accelerate damage,” Storm said.

Last year, EPRI completed a study examining 40 years of degradation data from the North American Electric Reliability Corporation Generator Availability Degradation System (NERC GADS) and found that boiler tube leaks were consistently the number one reason for failure in conventional steam plants. Increased cycling affects individual power plants differently—combined-cycle units are far more flexible by design than conventional steam units—but a more comprehensive view of the costs of cycling takes the following factors into account:

• More maintenance and overhaul costs—When plants cycle more frequently, the capital costs of maintenance and overhauls needed to avoid unexpected outages go up.
• Outage recovery—Getting units back up and running after a forced outage can be expensive. Those costs add up when cycling increases the number of unplanned outages. More frequent outage recovery can also introduce the potential for operator errors, increasing O&M expenses. Paying for generation to replace units during an outage is another expense.
• Increased heat rates—When units cycle more often and components degrade, there can be a long-term reduction in efficiency.
• Startup costs—Unit startup requires extra fuel, electrical power, and chemicals. When cycling demands more frequent startups, these costs add up.
• Shorter unit life—Ultimately, flexible operation demands shorten a power plant’s productive life, and utilities may need to pay for additional capacity.
• Engineering and management costs—Effective, flexible operation isn’t based on guesswork. Potential modifications and upgrades that allow units to cycle require study and analysis, which take time and financial resources. There are also costs associated with dispatching units to load to follow or take advantage of market opportunities.

Additional Inputs Deliver Additional Insights

The tool relies on a mixture of qualitative and quantitative inputs to project a forced outage rate and the costs of increasing the flexible operation of fossil units. For example, the tool incorporates historical data about a unit’s non-fuel O&M, capacity factor, heat rate, number of starts, operating hours, and generation. In addition, it integrates projections about a unit’s retirement date as well as the number of starts, operating hours, and generation.

Some of the inputs have nuance. For instance, not all starts have the same potential impact on plant components. “A unit may have been cold and sitting offline for two or three weeks, or it could be hot, and it just came offline yesterday because of economics,” Storm said. “The degradation and potential damage from those different starts are going to vary based on the operating mode, whether it’s cold, warm, or hot. And in some cases, hot starts may be more damaging to specific components than cold starts.”

Using the tool can give utilities some critical insights that can help guide future budgeting. For example, one output could be that each hot startup costs $41,000 in fuel and potential damage, while each cold startup costs $96,000. Those costs could then be used to forecast the annual expenses of starting a unit based on how flexibly it will need to operate. “You can look at a scenario where the historical cost of starts was $400,000 a year, but now it’s up 366 percent because of some change in the energy market,” said Storm. “Now the expected startup costs are going to be $1.8 million, which is huge if you only have an O&M budget of $2 million.”

These projected data can be helpful in making what are obviously complex decisions. For instance, imagine that a unit is supposed to be retired in five years and is being operated in a way that may lead to more forced outages. In that case, retiring the unit early may make more sense than paying for the necessary upgrades and maintenance.

The tool can also help decision makers grasp how proactive maintenance to prevent a forced outage can increase revenue. “Let’s say a utility does an asset integrity audit after using the tool and finds that it should fix some boiler tubes near the end of their useful life, or maybe it needs to upgrade its gas turbines,” Storm said. “It could then say to senior management that it needs $2 million to fix this. Then, if it has a summer or winter where demand spikes, it won’t have a forced outage that could cost $10 million in lost generating sales.”

In addition to the Cost and Uncertainty Management Tool, another high-level flexibility assessment tool is already in use in the United States, South America, Australia, and Asia. For example, India’s largest power utility, NTPC Limited, applied the tool at six of its power plants that have been particularly challenged by flexible operations. The plants account for 9 gigawatts of the utility’s total generating capacity of >55 gigawatts and have been tasked with operating more flexibly as more variable generation from renewables comes online. The tool has helped NTPC pinpoint flexibility limitations and prioritize investments that will enhance flexibility while maintaining reliability, safety, affordability, and environmental responsibility.

While it’s important to identify and quantify the potential costs of forced outages and the investments needed to avoid them, it’s also critical to understand how to address possible vulnerabilities. Another ongoing EPRI project will create a handbook of failure mitigation strategies to guide actions at individual units. The handbook will be released later in 2022.

“What we are doing is trying to optimize asset management under flexible operations. What are the failure mechanisms utilities are seeing? What is the frequency of those failure mechanisms? Are these short- or long-term issues?” Storm said. “We will take the understanding of issues or vulnerabilities and develop strategies that can address them in a practical way to enhance site defense strategies.”

---