Technical (Non-Financial) Benefits of FatiguePro
The advantages of fatigue monitoring have been acknowledged in a recent study on behalf of the US Nuclear Regulatory Commission where environmental effects on the ASME Code fatigue curves were being evaluated. It was concluded in this study that "… the best method to lower the cumulative usage factor for the few worst locations appears to be fatigue monitoring … in some cases, for example where thermal stratification exists, a combination of fatigue monitoring and more refined analysis may be needed."
Aside from the cost-saving benefits, the following reasons would drive a utility to consider fatigue monitoring with FatiguePro:
Technical Specification Requirements
Plant Technical Specifications require plant operators to count and categorize plant transient events. The purpose of this requirement is to ensure that the plant operates with the limits set by the design basis for the operating life of the plant (i.e., 40 years). Meeting this requirement demonstrates that the plant maintains structural margins. Although this can be done manually, the effort is greater and potentially less accurate than if an automated system is employed (especially if event-specific data is to be archived).
Cycle Counting Alone May Not be Adequate
For some components (like the feedwater nozzles), cycle counting alone may not be adequate for demonstrating long-term structural acceptability. Operating experience with all types of reactors world-wide has continually demonstrated that plants experience more cycles for some critical events than was assumed in the design basis. This makes it difficult to demonstrate acceptable fatigue when only the number of events is known. However, actual operating experience has also demonstrated that actual plant transients are significantly less severe than the events assumed in the design basis. Incorporation of plant severity thus overrides the fact that there may have been a higher number of events, and acceptable fatigue usage can be demonstrated. This assessment of actual plant events is best done with an automated system.
Plant Life Extension / License Renewal
Demonstration of acceptable structural margins (i.e., usage less than 1.0) beyond the original design life (i.e., 40 years) for license renewal most likely will require more detailed transient tracking efforts than cycle counting alone because of the 50% increase in the number of cycles (60 years of cycles vs. 40 years of cycles). As with #2 above, cycle counting alone may not be adequate to address the increased number of cycles considered for the license renewal period. This process is best implemented well before the end of 40 years to ensure that all issues are addressed by the time the license renewal period begins.
A baseline history of transient cycles and resulting fatigue usage can be derived after fatigue monitoring data is collected from one or two operating cycles. Cumulative fatigue usage since start-up can then be calculated based on review of available plant records. Then, all future transient events can be tracked and recorded along with the cumulative usage. This procedure establishes a strong technical basis for ultimately exceeding the number of cycles included in the original design. Alternatively, a partial cycle counting methodology can be employed to extend the life of certain components. In either case, actual component lifetime as affected by fatigue, can be projected.
Prioritizing Inspections
In some plant locations (typically Class 1 piping), there may be augmented inspection requirements (i.e., more frequent than once every 10 years) because of high fatigue usage. For example, the ASME Code has requirements for piping locations with a fatigue usage greater than 0.4. If it can be demonstrated that actual operating fatigue usage is much less than 0.4, then a technical argument can be made to remove some of the augmented inspection requirements (i.e., return the inspection frequency to 10 years). As with #2 above, assessment of actual plant fatigue is best done with an automated system, since every actual plant event is unique.
Increased Operational Awareness
Technical Specifications do not always give a completely accurate picture of what is going on in a plant. For example, in feedwater nozzles, thermal stratification is a relatively recent discovery, not considered in the original design. During Hot Standby, auxiliary flow initiated into the nozzle causes the interface level to cycle up and down with time. SI has investigated empirical data to develop a flow versus interface level correlation and a stratification stress matrix (stress for a given stratification level and circumferential location), which predicts axial stress cycling. The practical lesson learned here from FatiguePro is to shorten the duration of Hot Standby in order to protect the health of the feedwater nozzles. Without using fatigue monitoring based on real plant data, knowledge of the actual damage sustained by a particular location is limited.
The use of FatiguePro has also resulted in major discovery in PWR charging nozzles. The C-E CVCS design incorporates three positive displacement charging pumps and adjustable letdown capability. The consequence of this design is that during periods of letdown isolation with continued reactor coolant pump operation, a single charging pump is periodically started and stopped to maintain pressurizer level control. This design feature is common with all other C-E plants. The figure below shows one of these transients. A letdown isolation event spanning a ten-hour period occurred with associated charging flow isolations and restarts. The plant has a 6 gpm reactor coolant pump seal leakoff flow rate, which causes this previously undetected, near design basis (500F to 100F) charging isolation cycle to repeat 28 times, approximately every 20 minutes. As shown in the figure, fatigue usage climbed at the rate of 0.006 for each charging cycle during the letdown isolation. This ten-hour event caused a startling 0.016 increase in fatigue usage.
The relevant issue deals with the presumed exceedance of the number of Loss of Letdown and Loss of Charging events in the design basis. The design transient set included 100 Loss of Charging and 50 Loss of Letdown events. Plant staff state that neither of these plant events has been counted since plant startup, but that anecdotal evidence indicates that many letdown isolation events occurred prior to the installation of the letdown system accumulators in 1993, in some cases multiple events in a single operating shift. It is presumed with a high assurance that Loss of Charging events (the charging on/off cycles described above) would have also occurred on a 30-40 minute cyclical basis based on 4 gpm reactor coolant pump seal flow. It is thus reasonably certain that despite the lack of records, the design basis number of Loss of Letdown and Loss of Charging events has been surpassed.
However, experience gained by SI in the review of plant operating history using FatiguePro and other work performed for the plant indicates that a combination of 1,450 Loss of Letdown and Loss of Charging design basis events would still produce acceptable fatigue usage for the life of the plant (cumulative fatigue usage < 1.0).
FatiguePro offers much more information about what is actually going on in a plant and can provide "lessons learned" on how to minimize the fatigue damage your plant will sustain.
|
