By Anthony J. Carrino, Senior Consultant, HSB Solomon Associates LLC and Contributing Editor
The staff of most every power plant believes that their plant is “good” or at least “above average,” with the exception being those plants that have suffered with problems for so long that it has become painfully obvious that they could do better.
This universal “belief” begs the answer of just how good “good” is.
While many plants can be considered “good,” only a few can be considered “optimized.” That is, the trade-off between cost control, reliability performance and revenue capture and fuel conversion efficiency/emissions performance is managed in a way that optimizes the economic performance of the generating asset (or fleet of assets). Therefore, “good” by itself is insufficient to categorize performance. Knowing what any given generating asset is being compared against is one key to establishing better-than or worse-than average performance.
So what is meant by “optimized” in terms of a generating asset? Simply put, “optimized” means capturing the maximum net income possible on an ongoing basis while complying with legal, regulatory and corporate citizenship standards. Since most markets define roles for generating units depending on regional fuel mix, emissions limits, transmission constraints, technology mix and commercial or regulated market rules, “optimized” can be thought of as one or more generating assets achieving the best balance of sustained financial performance while minimizing safety and environmental incidents, given the market missions of those assets within the current set of constraints. Market mission success may be driven by high availability at all times (capacity payment-driven), while others only require high availability during key market seasons (peaking mission). Alternatively, market mission may be focused on being a low-cost provider while others depend on high thermal efficiency to compete against regional plants, or to minimize total costs or variable costs.
As such, consider the process by which generating plants are brought into sub-optimized conditions.
Reason 10. A simple thing like North American Electric Reliability Corp.-Generation Availability Data System (NERC-GADS) reporting impacts the financials of a power plant.
The NERC-GADS system of reporting generating unit reliability was established in the early 1980s as a response to blackouts. This voluntary reporting system was implemented to establish how much of the generating asset base is available for service to support the grid. Some generators still do not realize the usefulness of this reliability data in reallocating limited human and financial resources toward necessary betterment projects. Nonetheless, with the coming mandatory GADS reportingdriven by the Energy Policy Act of 2005this data reporting function will assume a higher level of importance for accuracy.
In addition to NERC-GADS data being used for the original charter of the system, it is being used as a basis of reliability-centered maintenance decisions and even for commercial market revenue calculations. Remember that in deregulated markets, these sorts of reliability indicators coming from GADS data are used to determine capacity payments for the coming year. If unreliability event reporting is in error, it could negatively impact installed capacity (ICAP)/unforced capacity (UCAP) calculations for the coming year.
(As an aside, GADS reporting is typically assigned to job functions as a training or business management function. Many times it is not reviewed by commercial or leadership personnel prior to submittal. Accurate GADS reporting will become increasingly important to identify emerging reliability issues as equipment and operating scenarios evolve, as well as knowing the commercial impact of how events affect revenue calculations.)
Reason 9. Capital requirements for environmental regulatory projects often impact planned maintenance activities by requiring deferrals of betterment projects or major maintenance spending that then become permanent maintenance spending reductions.
Compliance activities are always driven by date-certain compliance timetables, while plant betterment and major maintenance activities typically are not. Deferrals in plant betterment or major maintenance have finite lives before impacts are felt in production or reliability of the generating unit (research shows this to be in the range of two to four years). Understanding deferrals, mitigation activities and recovery plans is necessary to avoid temporary spending reductions on maintenance from becoming permanent reductions in maintenance spending and dropping reliability performance.
Reason 8. Operating and maintenance practices and procedures do not keep up with changing operating demands or staff turnover.
Examples of changing operating schemes for a given power plant asset exist as plant staffs struggle to realign operator training and operating procedures to current market conditions. Sometimes, as is the case with heat recovery steam generators (HRSGs), operating scheme changes have little immediate impact but rather a long-term effect that is not easy to quantify. As fuel prices, regional generating asset mix and emissions regulations continue to change, economics of generating plant projects will require continuous reassessment and operating schemes will be impacted.
In every recent industry conference, a common theme is human resource planning for ongoing and upcoming shifts in workforce demographics. Demands for skilled labor continue to rise as multiple sectors (including power generation) compete for scarce expertise. As new employees enter an organization, the lack of procedure reviews and updates for training programs are sure to contribute to performance impacts.
Reason 7. Comparative analysis is limited to internal comparisons of plants within a single company fleet.
While some companies depend on judging performance mainly by internal comparisons, taking a broader perspective on performance can provide leadership as well as plant staff with a larger sample size that cannot be represented by a limited fleet. When leadership sets goals based on known historical performance, plant staff believe the goals to be based in reality and, therefore, have more “buy-in” to these success measures. Broader samples allow generating units to be assessed with a sufficiently large statistical sample. If comparisons are made within a fleet with only one or two assets of a similar technology, fuel, utilization, or age, how much can be judged from that small sample set?
Reason 6. Comparative analysis and performance assessments are deemed “top plants” for a single attribute.
The “real world” of power generation includes a significant number of interrelationships between business decisions and generating unit performance. For a plant to be named a “top performer,” one should take into consideration a balanced perspective that recognizes the plant’s market mission as its primary economic driver, all the while keeping in mind key compliance measures. It is important to understand that there is no one solution to optimizing a generating unit; many methods have been implemented to solve the optimization equation. Focusing on a single performance measure tends to leave “money on the table” in other related areas.
Reason 5. Data is not used to make business plans in a changing environment.
If operating scenarios are changing or if emissions control modifications are required, a detailed database of real, historical performance and cost information can be useful in estimating the impacts from those changes. If a scrubber or SCR has to be installed, a complete database with tightly defined operating cost, removal efficiency, maintenance spending for routine and overhaul maintenance and reliability impact information can take the guesswork out of planning. Actual historical data can be used and adjusted for cost escalations or current market inputs.
Operational risk contributions to the overall generating unit can also be measured by sub-system, trended over time, and used to manage longer-term allocations of maintenance resources.
Reason 4. Past lessons and practices are applied without conscious thought or the realization that commercial drivers have changed.
Over our careers, we build our own “database” of practices that have worked for us in the past. While an emphasis on reliability of service was the highest priority in the past, knowing that more economical (and technical) alternatives may be available today depends on solid communication. Staying current on economic and market pressures are critical throughout the organization. The dialogue between plant operating staff and commercial staff is necessary to keep both sides well-informed of business drivers and generating unit performance capabilities and limitations.
Reason 3. Data is not used to estimate future performance.
Future reliability performance probabilities can be established from understanding the relationships of maintenance spending allocations. With data, testing future outcomes based on known, historical performance becomes possible. While not a guarantee of future success, large data samples of real assets can provide guidance to long-term spending plans for generating units. One example could be a plant that is equal to its peers in total maintenance spending, but which experiences reliability twice as bad as its peer. Understanding relationships within resource allocations can show higher or lower probabilities of reliability performance.
Reason 2. Normalization of differing assets based on utilization, fuel mix, technology and other factors is not understood or easily quantifiable.
Companies have struggled with the impact that cycling or intermediate load operation has on comparing performance among generating units. At the same time, as experienced generation professionals, we know that fuel matters, just as different technologies affect operating and maintenance costs. And of course, economics provide numerous examples of economies of scale and shared resources. Knowledge that these factors can be quantified with some precision eliminates the guesswork of normalizing the differences between a diverse fleet of generating unit assets by “gut” feel.
Reason 1. Performance measures are selected that allow some part of the generating unit operation to be “forgotten.”
All companies evolve in their thinking on setting performance measures. Where equivalent forced outage rate (EFOR) has been used in the past, not performing maintenance overhauls in a timely manner “gives up” some remaining availability that might be converted into net income. While “tested heat rate” performance measurement addresses recovery of the majority of design performance over time through major maintenance activities, understanding that starts, stops and derates for equipment failures and maintenance also affects the annual heat rate performance by requiring more consumption of start-up fuel (not to forget increased chemical and water consumption and even emissions allowances, as other forms of variable cost).
Similarly, focusing solely on O&M cost or controllable costs might leave out performance on heat rate, since O&M is usually figured as excluding fuel.
While some of these issues seem fundamental, most have been encountered in plants around the world, including in our own backyard. The best plants are not exclusive to the U.S. generating asset base. Excellent performers also exist in Europe, Asia and other parts of the world. Optimized generating units exist outside “traditional” power generating companies, too.
When goals are being set for your generating units, it must be realized that one generating unit cannot possibly be the lowest cost performer, the highest efficiency unit and the best at revenue capture by means of the highest possible availability all at once. This reality makes it necessary to choose one or two factors as primary and secondary goals, understanding that there will be tradeoffs in other areas in order to optimize the business. The difficulty in operating all generating assets with evolving market pressures is that all decisions include trade-offs and that is what makes the job so challenging for power professionals at any level of the organization.
Asset optimization can be achieved by understanding the total picture and communicating that to all employees within the organization. While additional performance goals can be set that are “controllable” at each level of the organization, the full picture needs to be well understood by the entire organization.
Setting performance goals in a constantly changing environment reflects the clear need to continuously reassess possible ranges of performance, given what we have to operate, the markets we are in and the need to utilize data to understand what is possible, or, to answer the question, “How good is ‘good’?”