Doing more with less: Leveraging your spatial asset data
Data - Sources, Target applications, errors & error correction
Sources & Targets
There are many sources of data needed by end-use Operations Support System (OSS)
applications. Many of these systems rely on a geospatial representation of network
assets. Accurate meter reading and other customer load related information, real-time
network element status, network connectivity, crew compliment and location, customer
status, construction order content and status, cost, and other basic information drive realtime
business decisions. Additionally, accurate and up-to-date detailed physical asset
attribute (size, rating, etc.) and configuration, customer-to-facility relationships, phasing,
electrical connectivity, construction configuration or type, and structural integrity data are
essential to maximizing the performance of many electric-based OSS applications.
Typical source systems include CAD, GIS, CIS, AMR, SCADA, Distribution
Automation/SCADA, Asset Management Systems, and Graphical Design Estimation.
The data from these systems, in turn, drives Outage Management (OMS), Work
Management (WMS), Distribution Management (DMS), Energy Management (EMS),
and Engineering Analysis packages. The conjunction of data from such a wide range of
sources opens the door to data omissions, conflicts, gaps, and errors.
Errors
What types of data errors can impact an OSS and where do the errors come from? Data
errors can result from data migration and data integration efforts if these projects are not
properly monitored and controlled. Here, incompatible data formats, incomplete data,
and duplicate data from two or more sources can offer stiff project challenges above and
beyond what may have been inaccurate to start with. However, many errors result from a
lack of proper data-entry business processes and data ownership. Typically, these errors
arise due to limited error checking at various stages in the data management process.
Another contributing factor is the inability of personnel who rely on the data to easily
make corrections. Other error sources are often beyond the reasonable control of the
utility. For example, when performing restoration efforts after a major storm (ice,
hurricane, etc.), the primary focus is to get power back to customers, not to accurately
report how the network may have been altered. Data-capture problems during these
emergencies are compounded by the use of “foreign” crews who are not familiar with the
local utility’s data-capture procedures.
The following represents the situation encountered after performing an initial assessment
on a typical sample data set:
- Anchor Text Outside of File (39 of 7,992)
- Linear Objects (Primaries, Secondaries, Services) with 1 vertex (1,582 of 26,682)
- Construction Spans w/o phasing (132 of 3,174)
- 30 missing phasing
- 102 missing phasing, node ids, and circuit ids
- Fuses/Switches/Reclosers w/o conductors (348 of 348)
- 102 not aligned w/ conductor or Open/Close status unset
- Primary Conductors node id unset (6 of 20,513)
- Network connectivity issues (132 of 3,193)
- 75 logical connectivity (node id attribute) problem
- 57 graphical connectivity problems
On a percentage basis, these problems may not seem extreme, but their impact on key
operations support system applications can be very significant.
The connectivity and phasing problems discovered would have a severe impact on the
operations of an OMS or an engineering analysis package. Inaccurate data can create a
multitude of problems—wrong assumptions about which device has operated to clear a
fault, thus resulting in sending crews to the wrong locations, over- or under-designed
facilities, and false facility loading predictions that lead to shortened facility life and
potentially dangerous situations.
Other errors are associated with switching updates, loads, construction types, conductor
sizes, consistency in device information like regulator settings, and customer-to-facility
(usually the transformer) linkage. These data errors can also pose problems for a variety
of applications. Fuses won’t clear faults properly and cause damage to expensive utility
and customer assets. Upstream isolation devices may operate before downstream devices
have a chance to isolate a fault. This leads to interrupting more customers than
necessary, negative impact on performance metrics, an increase in customer complaints,
and possibly to performance rate penalties. So, the question that remains is how do we
detect and fix these errors?
