LER screening algorithm for identification of potential accident sequence precursor events Page: 4 of 6
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
3. System and Train Level Performance. The system
involved (e.g., feedwater system, high pressure coolant
injection system), the number of trains involved (if
applicable), the cause and mode of failure, and the timing of
the failure (e.g., pre-existing, potential, or actual failure) are
4. Effect on the Unit or Environment. The effect on the
unit includes reactor trips or shutdowns, power reductions,
engineered safety feature actuations, actuation parameters,
types of actuations, and initial plant conditions. The effect on
the environment includes radiological releases or personnel
Each LER sequence will consist of one or more of each
of these basic step types which are linked together in a
searchable, time-ordered sequence or sequences to form the
entire event. As of mid-1996, the database contains
approximately 43,000 LERs with event dates from 1980 to
within about 60 days of the current date-LERs are added to
the database within about two weeks of when they are
received at ORNL.
The database supports simple and broad searches, such
as identifying LERs at selected plants that involve certain
component types. It also supports far more detailed and
complex searches, such as identifying events where a design
error in the feedwater control system led to a steam generator
low level trip and where one or more trains of AFW failed
subsequent to the trip due to a procedure-related operator
error. Many different types of these conditions may be
combined with Boolean operators to further define searches.
The ASP algorithm is one of the more complex SCSS search
III. ASP MANUAL SCREENING
ASP screening has been performed by experienced
analysts who often review numerous events consecutively.
Events with elements typical of precursors are set aside for
detailed analysis and quantification. Manual screening is
normally very effective for a small number of LERs, but it
becomes less effective as the number of events increases. The
events must be carefully studied to catch incidental remarks,
especially concerning unexpected system responses or
equipment unavailabilities. To ensure that no potential
precursors were missed, each LER was reviewed by two
engineers. Approximately 1,200-1,500 LERs are reported
each year, therefore, a considerable amount of effort was
expended to review the events with recognition that
effectiveness in identifying the events drops as more have to
IV. SCSS COMPUTERIZED SEARCH STRATEGY
An SCSS screening algorithm was developed to address
the intensive review activity deemed necessary to identify all
potential precursors over a yearly period. The purpose of the
algorithm was to reduce significantly the number of LERs
subject to detailed review by ASP project staffyet still identify
all potential precursors reported in LERs.
The algorithm capitalizes on the intensive LER review
and encoding already used for SCSS and utilizes SCSS'
extensive search capabilities. The algorithm was constructed
in a manner analogous to the ASP manual screening. The
algorithm requires that (1) one complete train of two or more
safety systems or (2) one entire safety system is faulted or
unavailable. It also includes the occurrences of initiating
events such as LOCAs or LOOPs or reactor trips with one or
more trains or an entire safety system unavailable. Several
search groups based upon the conditions above are then saved.
Set algebra is then performed on the groups in order to
prioritize the groups having the highest likelihood of
containing precursor events. The groups are:
1. potential LOOPs and LOCAs;
2. reactor scrams plus faults in at least two safety system
trains (e.g., a scram and a fault in one train of two
different safety systems or a scram and one entire safety
3. faults in the refueling water storage tank involving
multiple instrument faults, multiple heat tracing faults,
leakage from the RWST, interfaces with the residual or
decay heat removal (RHR) or chemical and volume
control (CVCS) systems;
4. reactor scrams plus a fault in one entire safety system;
5. faults in trains from at least two different safety systems
or two entire safety systems faulted;
6. faults of the entire emergency power system; and
7. other safety systems faults (not including the emergency
The algorithm was executed on the SCSS database and
compared to results from past manual ASP reviews for eight
years. Several significant benefits were realized. First, the
algorithm provided a reduction of approximately 75% in the
number of LERs that were required to be reviewed and
identified about 98% of the precursors during the years of
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Poore, W. P., III. LER screening algorithm for identification of potential accident sequence precursor events, article, September 1996; Tennessee. (digital.library.unt.edu/ark:/67531/metadc672960/m1/4/: accessed April 24, 2018), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.