Electric utilities search fo ways to improve service reliability to their customers. PPL Electric Utilities Corp. (PPL EU, Allentown, Pennsylvania, U.S.) is no exception. It added troublemen off-hours, staggered shifts during daylight savings time and automated critical-distribution circuits. Recognized as a Tree Line USA utility for five years in a row by the National Arbor Day Foundation, PPL EU also determined the optimum tree-trim cycles for more than 27,500 circuit miles (44,257 km) of overhead distribution circuits, as well as more than 4500 miles (7242 km) of transmission rights-of-way (ROW) corridors.

TRIM-CYCLE HISTORY

PPL EU delivers electricity to about 1.4 million customers in central and eastern Pennsylvania, and topographically and demographically, its service territory is diverse. Conditions range from the farmlands of rural Lancaster County to the densely wooded Pocono Mountains, to the urban centers of Allentown, Harrisburg and Scranton.

Prior to this study, PPL EU adopted distribution trim cycles of eight years for circuits in rural areas (average customer density of fewer than 35 customers per circuit-mile) and five years for urban circuits. The utility places urban circuits on shorter cycles, because many of these lines were built where the full desired clearances could not be obtained because of homeowner concerns or ROW issues. An annual prioritization process identifies circuits that have had multiple tree-related outages. Other indicators also help determine the order of circuit trimming. The resulting mileage of planned distribution trimming averaged about 4100 miles (6598 km) per year.

For PPL EU, outages from trees are a major cause of sustained customer interruptions. Like other utilities, PPL EU classifies these outages as either caused by inadequate trimming or by trees outside the maintained ROW. Outages from trees outside the ROW outnumber those caused by inadequate trimming. PPL EU's trimming strategy includes the identification and removal of hazard trees outside the ROW for which landowner permission can be obtained. However, because of the densities of trees in certain parts of the service area and the sometimes capricious nature of trees that do fail, it is not possible to identify and remove all potential hazard trees.

RELIABILITY IMPROVEMENT

PPL EU adopted a series of initiatives to reduce the number of customer minutes lost due to interruptions (SAIDI) by reducing the frequency of sustained customer interruptions (SAIFI), reducing the duration of those outages (CAIDI) or reducing both. The company's goal is to achieve first-quartile performance among its peers, and management is willing to fund additional maintenance and capital projects as long as they are cost effective. Vegetation management was already one of the highest-funded and most visible maintenance programs. PPL EU studied how to reduce the trimming cycles of distribution circuits to identify cost-effective initiatives to improve SAIDI.

The approach taken was similar to other studies previously completed on PPL EU's other maintenance practices for its distribution, substation and transmission systems. The company established cause-and-effect relationships between about 18 different maintenance programs' costs, contributions to reliability and cycle times. The study focused on how to develop similar relationships between the distribution-trimming cycles and reliability, based on whether the circuit was classified as either rural or urban, and whether the outages were caused by trees from within or outside the ROW.

APPROACH

PPL EU collected decades worth of field data on sustained outages and tree-trimming cycles. The analysis began with a comprehensive review of the existing databases, which data were relevant to the study and how best to link them to achieve a single, smaller set of observations for analyses.

The outage database had data tracing back to before 1992, as did the tree database. PPL EU decided that 15 years of data would be sufficient for the analyses. The outage database provided data for every sustained outage (defined as a customer interruption lasting 5 minutes or longer) including its length, cause code and number of customers affected. PPL EU's outage management system automatically generated this data. The tree database captured information from the billing records of contractors, who perform all of PPL EU's vegetation management services. From these data, the utility was able to determine the year in which the circuit was last trimmed, the total length of the circuit at that time and its urban or rural designation.

Tied together, these databases allowed the generation of thousands of observations that tied each tree-related outage to the nature of the circuit (rural or urban), the length of the circuit (to normalize the data to account for different circuit lengths), the contribution to SAIDI, whether the cause was due to inadequate trimming or trees outside the ROW, and the number of years prior to the outage that the circuit was previously trimmed.

Few outages occurred one or two years after a circuit was trimmed as a result of inadequate trimming. In addition, the number of outages 10 years after trimming was also small because a circuit rarely exceeded that time span, especially if that circuit had a history of tree-related outages.

RESULTS

The results of the analyses provided several insights into the effects of the distribution line tree-trimming program. The top figure shows the failure curves that were determined by the study, grouped by cause. The X axis is the number of years from when the circuit was last trimmed. The Y axis is SAIDI contribution. The lower curve is outages due to trees within the maintained ROW. These curves show the very strong positive correlation between trim cycle and SAIDI contribution. It also shows that changes to the tree-trim cycle will affect both causes similarly. Outages caused by trees outside the ROW are reduced with shorter trim cycles due to more frequent inspections of the lines and identification of hazard trees.

The bottom figure shows similar curves, with the lower curve showing outages on urban circuits and the upper curve rural circuits. These curves demonstrate that SAIDI would be significantly reduced by shortening the trim cycle for rural circuits and would be reduced, but not as significantly, for reductions in urban cycles.

RECOMMENDATIONS

Based on these curves, the preliminary recommendation was to reduce rural trim cycles to five years and keep urban trim cycles to five years. The expected improvement in system SAIDI was 16 to 18 minutes by 2011 for the rural cycle reduction. The total number of trimming miles would increase about 30% to nearly 5400 miles (8690 km) per year.

This preliminary recommendation first went before PPL EU's forestry professionals who manage and direct the work performed by contractors in the field. They concluded that because of the ability to achieve better clearances in the rural areas, a six-year cycle was a better choice. The change in SAIDI between years five and six is minimal. The foresters also recommended going to a four-year cycle for the urban circuits because of the inability to achieve desired clearances. The improvement in urban-cycle SAIDI offsets the change in rural cycles, so that the net result was essentially the same for both reliability and total miles trimmed per year.

The senior management approved this recommendation, as well as the funding necessary to achieve the mileage required. PPL EU managers added these funds from new sources, not by cutting back or reallocating funds from other maintenance activities.

PROGRESS

The ultimate results of this initiative will not be readily apparent until an entire trimming cycle has been completed, which will be 2011 at the earliest. However, customers on those lines trimmed earlier in the cycle should see improvements within the next few years. The cost for PPL EU to achieve this SAIDI improvement compared well with other initiatives being undertaken. Employee and contractor morale is improved due to the visibility and support the distribution vegetation-management program has gained from senior management. Also, this analysis gives PPL EU defensible arguments on optimum trim cycles that balance reliability, and ultimately the costs to our customers. As this initiative goes forward, the utility will be closely monitoring its effectiveness.

Philip F. Weber is the system maintenance engineer for PPL Electric Utility's distribution system, responsible for the reliability and maintenance of overhead and underground distribution circuits, including the vegetation management program. A 30-plus-year veteran of the electric utility industry, Weber earned both his bachelor's degree in industrial engineering and a master's degree in management science from Lehigh University in Bethlehem, Pennsylvania, U.S. He is a registered professional engineer. pfweber@pplweb.com

Jonathan Tillson earned his bachelor's degree in integrated business and electrical engineering from Lehigh University in May 2007. He is pursuing his master's degree in analytical finance at Lehigh. jht2@lehigh.edu

Pertinent Data

OMS Outage Data	Tree Data
•Date of outage	•Circuit
•Time	•Date trimmed
•Duration	•Circuit length
•Cause code	•Rural/urban
•Number of customers	•Contractor
•Circuit	•Cost
•Fault location