Age Concern1 April 2000
Operating and maintaining ageing nuclear power plants with life expectancies of up to 40 years presents a number of challenges to the I&C engineer.
British Energy operates eight nuclear power stations: seven are twin AGRs, and the eighth is a PWR. The I&C design for the AGRs started in the late 1960s and the stations were commissioned progressively from the late 1970s to the late 1980s. The PWR I&C is largely based on 1980s technology, but all the AGRs still have a large amount of the I&C they were built with, which has been carefully maintained and in some cases replaced.
I&C equipment has a finite lifetime, which in some cases is less than the working lifetime of an AGR/PWR and licensees must expect to replace some equipment. Rapid developments in computing and the increased use of semiconductors have resulted in a fast moving marketplace, and many products become obsolete well before the end of their useful life.
In the case of British Energy and its suppliers, demographic changes such as staff movements, and changes in the staff age profile are also a concern, as they affect the levels of expertise that are available to provide system support.
Obsolete I&C systems can be maintained, with an adequate supply of spares and maintenance expertise on hand. Where replacement is considered, the licensee needs objective information about the current state of the plant, so that he can make proper comparisons of the need for capital investment across all areas of the business.
In England, the BE reactors are operated by British Energy Generation Ltd (BEGL). The company needed a strategy to manage its I&C systems over the plant lifetime to ensure that they continue to operate safely and reliably.
The strategy had to incorporate obsolescence, ageing and maintainability. The US-based research consortium, Electrical Power Research Institute (EPRI) has developed a structured family of methodologies for the identification and resolution of I&C lifecycle issues in the nuclear power industry, known as the EPRI I&C Upgrade Initiative. BEGL had access to the methodology via its membership of the EPRI research programme.
The problem facing BEGL
Several different tools are available for tackling individual aspects of the I&C lifecycle problem. EPRI has developed a family of methodologies that will address all aspects of the problem in a systematic way, to ensure that:
• Plant operators target resources where they will be of most benefit (priorities are based on safety and reliability) and all methods are based on a cost-benefit analysis of recommendations.
• Operators identify I&C systems for retention or replacement as appropriate based on a defined planning horizon (usually 10 years), requiring the production of a system maintenance plan (SMP) or an upgrade evaluation (UE), respectively.
EPRI has identified four methodologies:
• Lifecycle management plan (LCMP) methodology.
• Plant communications and computing architecture plan methodology (PCCA).
• System maintenance planning methodology (SMPM).
• Upgrade evaluation methodology (UEM)
The methodologies are applied in the following way. A LCMP will be produced for a station, or for a subset of systems on a station. One important factor that is set at this point is the period that the plan will cover. The LCMP baselines the current I&C equipment to identify candidate systems for either upgrade or retention, and to provide a means for prioritising the work to be carried out. Systems are likely to be candidates for an upgrade where continued maintenance is impractical for a combination of obsolescence issues, or where a significant improvement in plant performance can be achieved to offset the cost of new equipment. However the methodologies recognise that retention is likely to be the most cost-effective choice for short and medium term planning periods.
Where systems are identified for retention, an SMP is developed to look at maintenance issues that will affect the system during the plan period. The SMP guides the user to determine the root causes for maintenance-related problems, and then to carry out cost-benefit analysis to rank the equipment and process improvements identified by the analysis. British Energy has found that where systems are to be upgraded over several years, a SMP will be produced in parallel with the UEM to ensure that maintenance problems on the old equipment are managed through the upgrade period. Issues that are covered during the production of the SMP come under the broad headings of obsolescence, reliability and maintainability. A key factor for the company is the availability of suitably experienced people to work on its system hardware and software. This helps to identify further staff and training needs. The SMP is a ‘living document’, which must be subject to periodic review.
For upgrade candidate systems the UEM is followed. This allows system performance to be reviewed at a higher level than is achieved through the SMP process. The product of the UEM is the upgrade evaluation report (UER) which sets out to examine the current performance of the system, and then looks at modifications that could be made to enable it firstly to meet, and then to exceed its existing requirements. Conceptual designs are proposed at this stage and cost benefit analysis undertaken. The document would conclude with a recommended course of action. It may conclude that an upgrade is not required for the period under review. In this case material from the UER will be used as the basis for an SMP.
BEGL has applied the methodologies flexibly. For example, the company does not always produce an LCMP for the whole station. Instead it may produce either the SMP or UER for a system that obviously requires attention, or a LCMP for a set of subsystems. It has also worked with contractors to develop abbreviated versions of the methodologies.
Life cycle management
BEGL’s I&C legacy developed over the last twenty years. During this period the plants passed from the public sector (part of the Central Electricity Generating Board), through privatisation into National Power and then Nuclear Electric, before finally arriving as part of British Energy. As the plants are now part of a smaller group, concerned only with nuclear generation, life cycle management now has a clear focus.
Since the plants were put into service, there have been a number of significant modifications to the I&C and data processing (DP) equipment. Work includes:
• An Argus 700 emulator developed for the Argus 500 data processing system.
• Refurbishing boiler controls.
•Adding overview panels to original control room desks to aid operators during high alarm activity and loss of main computers.
• Installing microprocessor-based turbine governor equipment.
• Replacing DP systems (see article on Hinkley Point B, following).
• Adding a diverse protection system.
There have also been minor refurbishments. All these modifications have helped to extend the life of the I&C equipment. While in some cases obsolescence was the key driver, some came out of a need to extend and improve performance. The modifications were not part of any integrated strategy.
Despite the work carried out to date the plants still have much of the I&C they were commissioned with. In many cases this equipment is bespoke, obsolete and no longer supported by the original manufacturer. As a consequence BEGL is now conducting a wide spectrum of I&C lifecycle management studies, ranging from projects developing SMPs for single pieces of equipment to developing a LCMP which will cover a range of systems.
In non-nuclear generation, I&C system replacements and upgrades are possibly considered routine and the obvious solution to life cycle management problems. But in nuclear generation, procedure requires a rigorous examination be made of the safety implications of making any modification to a plant. Once a safety case has been established, based around a particular arrangement of equipment, a licensee needs to review any deviation from this arrangement for nuclear safety implications. This has led to a close adherence to established arrangements for safety critical applications and reduced the scope for taking advantage of some modern technologies. Furthermore, any changes in equipment and procedures will require retraining of operational staff, and these costs must be taken into account in the total picture.
Engineering in this environment puts special pressures on BEGL and its suppliers. The company’s experience of the supply chain to date is:
• Equipment becoming obsolete at an accelerated rate.
• A legacy of bespoke and in-house solutions.
• Maintenance of key skills becoming an increasing overhead.
• Suppliers in general providing a poor level of long-term support.
• An increasing reliance on a small number of specialist suppliers.
• Difficulty in applying component reliability information, especially to predict end of life.
• A shortage of suppliers who can reverse engineer their way to improved safe and cost effective solutions.
The EPRI I&C lifecycle management strategies have provided BEGL with a structured method to address or handle these problems. This has led to:
• Staff retraining.
• Lifetime purchases of key spares before they become obsolete.
• Small size replacements to free up spares for other parts of the system.
• Pressurising suppliers to refurbish or reverse-engineer equipment.
• Special production runs of spares.
• Where no other solution exists, larger scale replacements.
When replacing equipment, the company places greater emphasis on life cycle management aspects at the specification stage. It treats obsolescence management as an important part of design and criteria when selecting a supplier.
Control room computers
BEGL’s largest I&C life cycle study started in summer 1999, and involved a DP computer providing reactor control and control room displays. The study is taking in ancillary computer systems, with a view to ensuring that they too will support station life. If any replacement work is commissioned, it may be possible to rationalise the way these systems are used and information is presented to the plant operator.
In this case, BEGL is producing the SMP in parallel with the UER. The company has adopted this policy because it believes it will have short-term maintenance issues to resolve while addressing wider issues identified by the UER.
This study covers a more modern main DP, which comprises two computer systems, providing real time data acquisition, control, display and data logging functions to support operation of the two reactor/turbine generation units from the central control room (CCR). The station also has a training simulator, which employs a similar computing system design as the DP.
For this project the objectives have been defined in terms of managing the lifecycle on a planned basis to:
• Improve safe operation. The DP provides plant information and alarms in the CCR; it contributes to safety and environmental monitoring.
• Grow profitability. DPS availability is essential to maintain generation. Obsolescence must be managed to minimise costs over the station life.
• Manage with limited resources by replicating company best practices and adopting industry standards.
• Provide a quality service to stakeholders. Determine the need for DP replacement or maintenance, and provide the optimum life cycle strategy.
The study has now been running for just over 18 months and a UER has been produced for the systems. BEGL has asked a number of suppliers to carry out a design study to examine alternatives for the replacement of the DP man-machine interface (MMI).
Within the UER, tables provide a summary of the system problems, root causes and corrective actions identified as part of the system analysis. Most of the problems are associated with obsolescence. The system components have therefore been assessed in terms of severity, to forecast when they may become a threat to station operation. (This approach is different from the EPRI methodology, which recommends that problems be assessed in terms of payback period).
A range of corrective actions may be needed for each problem, from quick-fix repair, to refurbishment or replacement options. The choice of corrective action will depend upon the implementation timescale and the expected extension in system life that the corrective action will provide. For example, it may be possible to implement a quick-fix repair within 12 months, extending the life of the system by three years, whereas a replacement may require five years to implement but achieve a ten year life extension. Estimates are made of the time required for implementing a corrective action and the life extension expected from the work.
In planning replacements on an operational power station, a baseline requirement is to implement replacements in a manner that will ensure safe operation without interrupting plant operation. It is also important to minimise the changes experienced by operating staff. Safety and operational constraints mean that corrective actions may need to be phased and that a combination of both short-term action to extend the system life, and long-term corrective action (replacement) is appropriate to manage the DP lifecycle over the remaining station life. Therefore, when scheduling the timing of implementation it may be wise to start some work earlier than necessary.
Three identical design study contracts were placed with suppliers of DP systems under a common specification to investigate feasibility, best practice, cost and programme for a replacement MMI system.
BEGL’s objectives for the design studies were to establish from a suppliers viewpoint:
• The feasibility of an MMI replacement.
• A baseline for best engineering practice for replacement design implementation and ongoing support.
• Areas of threat.
• A budget for a potential replacement.
• The level of compliance and fitness for purpose of specific proposals.
• The outcome of the design studies will serve to substantiate and refine the proposed approach for replacement of the MMI parts of the DPS. The results of this will be used to update the concept design for replacement of the MMI documented in the UER.
The design studies have confirmed that the design intent for MMI replacement is feasible.
The design studies have identified the need to ensure the portability of the application software over the station lifecycle. However, long term support costs will be significantly reduced by adopting a system design based on open system architecture.
Reactor safety circuits are a priority for evaluation. Their channels can be compared easily because they have the same reliability and licensing requirements. It is a longer term objective to consider LCMPs for other I&C systems and equipment.
This LCM includes firm recommendations for retained systems and upgrade candidate systems. The list of retained systems is prioritised to indicate the order in which SMPs are prepared. The list of upgrade candidate systems is prioritised on a preliminary basis only for production of evaluations.
The EPRI method is a workable means of monitoring the state of the reactor safety circuits and determining which, if any, should be considered for upgrade. Because much of the equipment is of similar age and design, the company introduced a method of scoring to compare system performance and to confirm upgrade candidates. This gives a degree of independence to the results. It can also be reviewed and kept up-to-date with comparatively little effort.
Three principal areas are addressed by the EPRI methodology: reliability, obsolescence, and maintainability. There is evidence of issues that need to be addressed in all three areas for the reactor safety circuits.
The plan concludes that some systems should be classified as upgrade candidates. Partial or complete replacement should be considered in detail and subjected to cost-benefit analysis to determine if they are sustainable. High equipment failure rates, combined with rising trends of failure and maintenance effort, are the principal reasons for considering these systems for upgrade. By comparison, other safety systems do not have this combination of high failure rates, which are steadily increasing in level. These systems also have important obsolescence concerns.
For the retained systems, some transducers have historically had high failure rates, but they are not increasing. For this equipment, apart from monitoring future failure rates, no action is necessary. However, in other cases there is a significant rising trend of corrective maintenance, as judged by the station work-order time bookings.
BEGL will review spares holdings for all equipment, which it will maintain in its existing form for the foreseeable future, to ensure that adequate levels, particularly of obsolete components, exist for the planned period. Future tracking of trends for the flux systems from data held on information management systems would benefit from recording work order details against specific plant slot references, rather than against general numbers, which occurs at present to some degree.
Key to the BEGL strategy is the EPRI I&C Upgrade Initiative and this is being applied across a number of stations. The key drivers behind the BEGL strategy are maintaining safe generation, extending plant life and demonstrating a return on investment. In BEGL this is likely to mean a move away from bespoke systems and toward open platforms.
Due to the nuclear safety implications involved with making modifications on nuclear power plants, the company’s preference is likely to remain searching for innovative solutions to I&C lifecycle management problems. These solutions will involve training, reverse engineering, refurbishment and replacement only where it can be justified in the interest of safety and commerciality.