Fossil power plant residual life optimization in the USA
In the late 1970s and early 1980s US electric utilities were faced escalating costs and slowdowns in adding new capacity of all types. Nuclear programmes were scaled back in the face of safety concerns and increased regulations, while fossil plant construction lagged due to higher interest rates and environmental pressures. This created strong pressure to continue the operation to older units. Now, however, this emphasis is on unit economic optimization and plant profitability. This theme was developed in a greater detail by Tony Armor in his keynote address to the Ibeldrola Life Extension Seminar in Bilbao, Spain on 16 February 1995.
Based on current of equipment aging phenomena cannot operate safely for 50 to 60 years longer even though nearly half of all plants have traditionally been retired before before their fourtieth year of service. However, preparing units for service beyond historic retirement dates requires careful system planning to select the appropriate units for continued and the development of the new engineering methods for estimating the remaining useful life of service-exposed components.
The term “life optimization” is a more accurate description of the present utility approach to what was previously termed “life extension”. It recognizes the economic realities of the steps needed to keep an old fossil plant in service. In effect, the term underlines the “value-added” for each invested dollar. And in an increasingly competitive world, all operating units will by justified on the basis of generating a kWh as well as on the need capacity.
At the same time, the utilities are conscious of shrinking reserve margins and are aware that if older plants can’t supply the power, it has to be found elsewhere. This paper seeks to put the utility dilemma in perspective. It reviews ten years of “life extension“ activity in the Us in a climate dominated by environmental legislation and competitive forces, and reassesses the best approach for handling and aging fleet of fossil plants.
Utility industry changes
Deregulation and compettion are causing a fundamental reassessment of utility generation. The process of buying fuel, converting it to electricity, and selling it to customers in a specific service area is a well defined process that utilities have done well for nearly 100 years. Now othersare moving into each of these three activities with their own assets (mainly new power plants), causing utilities to re-evaluate the comparative worth of generating units that may be 40 years older.
Utilities often see better opportunities noe in wheeling power through the system to other utilities, or in maximizing profits from fuel procurement activities, or in building small plants in service territories belonging to others, or even in diversification ventures unrelated to electricity production. In such situation it becomes even more important to make overall assessments of existing power plant worth and likely future operation.
In the early 1980’s many utilities were considering life extension costs of around $300 – 400/kWh to keep the older plants operating reliabely. These estimetes were based on the possible replacement of many major components within the power plant. However, bwcause of the adoption of sensible residuallife assessment approaches and integrated corporate planning activities, most of these replacements and the associated costs have not occurred.
The assesment of component condition for fossil-power plant continues to be of major importanceto utilities worldwide and is clearely the prime focus of utility life optimization programmes. Figure 2 shows clearely that the reliability of power plant equipment significantly decreases after 20 – 30 years of service, so autility programme of planned maintenance is critical to economic operation of the plant.
Utility strategies
The reality of the “old plant” dilemma in the United States is that the following issues exist at the same time and, from an economic viewpoint, require an optimized approach:
1. Not enough new plants are planned in the next 10 years to reserve margins, even if no fossil plant capacity is retired. This is likely to lead more short-cycles options, such as orders for gas turbine based plants.
2. Old plants, greater than 30 years in age, can be operated indefinitely, but at progresively lower availability as maintenance demands increase.
3. Old plants are at risk when pitted against new constructions since availability and efficiency are likely to be lower. They will also be called for more cycling duty, further eroding these performance parameters. Therefore, utilities will be vulnerable to competition by aggressive IPP ’s and co-generators for prime industrial loads in the new competitive US environment.
4. The utility, in planning a life optimization strategy, must walk the fine line drawn by environmental legislation. In particular, a maintenance plan which calls for environmental clean-up equipment, such as scrubbers, can more than double the capital investment required for continued operation.
5. All up-grade decisions must be made in an environment of technical uncertainty. Decisions to replace, or not replace, key pieces of equipment can have far-reaching consequences in terms of future reliability. This implies greater dependency on state-of-the-art live estimation methods and modern decision making theory.
ON-LINE DIAGNOSTICS MONITORING
To achieve economically viable availability levels in older fossil fuel units, traditional maintenance practices are changing. A new set of maintenance practices, based on on-line monitoring techniques, as well as advanced off-line inspection and life assessment, can help utilities determine the precise condition of key components and systems.
Mounting evidence from the first wave of monitoring systems in the US (figure 3) supports the contention that by detecting and diagnosing abnormal mechanical behavior at an early stage, utilities can reduce plant downtime. Plant engineers and operators can use information provided by on-line monitors to avoid a forced outage and instead schedule the repairs.
Maintenance personnel can use this information to determine what maintenance to perform, and to ensure that spare parts, tools and manpower are available when the equipment is taken out of service. In addition, on-line monitoring reduces plant downtime and labour costs during a scheduled outage because plant personnel can tell when components are not in need of maintenance.
Modern diagnostic systems for fossil and nuclear power plants are generally based on microprocessor systems, which automatically track parameters of interest and flag abnormal situations. Often, the investment in a specific diagnostic monitoring system is easily offset by the avoided costs of one unanticipated outage.
To demonstrate the full potential of the available and emerging monitoring and diagnostic technologies, EPRI and PECO Energy (Philadelphia) have integrated diagnostic-monitoring system into a predictive maintenance programme for boilers, turbine generators, environmental controls, and balance-of-plant equipment at PECO’s Eddistone station (figure 4) as described in MPS, May 1991.
With virtually every major plant component permanently wired with sensors and feeding continuous data through a fibre-optic highway, operators and maintenance engineers at Eddistone are able, in real time, to observe the actual effects of operation on the performance and wear of critical parts and subsystems.
The signals from several computer-monitoring systems are integrated for display and analysis in a predictive maintenance diagnostic center that addresses operator needs. Special hardware and software interfaces have been designed to bring the data from incompatible computer operating systems under a common set of diagnostic display terminals.
Utilities in the US have embraced diagnostic monitoring technology and EPRI gas documented utility benefits exceeding 11$ million/year for member utilities.
Advances in life assessment
Life estimation is the crux of the whole plant evaluation exercise. It must provide estimates of creep life and cyclic life used to date; residual useful life remaining; operating parameters, which can lengthen remaining useful life to delay major replacements, and finally a replacement schedule which can be integrated within future operation and maintenance policy.
This activity suggest when major expenses in replacements are required, leading to a phase approach to life extension. Typical damage mechanisms are shown in Table 1.
