Arboriculture & Urban Forestry 37(4): July 2011 Arboriculture & Urban Forestry 2011. 37(4): 147–151 147 Tree-related Electric Outages Due To Wind Loading Siegfried Guggenmoos Abstract. In the aftermath of a wind storm that interrupted service on more than 40% of Puget Sound Energy’s transmission system, the regulator or- dered an investigation to evaluate options for hardening the electric system. The initial phase of the study quantified the extent of tree exposure along the transmission system and examined the correlation between various field measurable variables and tree-caused outage frequency. This phase of the work correlated 10 years of wind data with 10 years of outage data. Tree failure rates were calculated and used to model tree-caused outage expecta- tions based on wind speed. Models were developed to assess the outage impact of interventions that reduce the transmission system exposure to trees. Key Words. Electric System Reliability; Reliability Modeling; Tree–Conductor Contacts; Tree Exposure Risk; Tree Risk Quantification; Tree-related Outages; Storm Hardening; Utility Arboriculture; Wind Loading. Puget Sound Energy (PSE) is a utility providing elec- tric and natural gas service in western Washington, U.S. PSE has about 17,300 kilometers of electric distribution lines and approximately 3,360 kilometers of electric trans- mission. For the purposes of this study, transmission re- fers to lines energized at 55, 66, 115, and 230 kilovolts (kV). In December 2006, Puget Sound Energy experienced sub- stantial electric system damage and service interruptions due to gale force winds. Soils were saturated from previous rains and tens of thousands of trees uprooted and fell into power lines and substations (pers. comm. PSE, Hanukkah Eve Wind Storm of 2006, Wikipedia). The 115 kV system was particularly heavily impacted. In the aftermath of this storm, the Washington Utili- ties and Transportation Commission, or regulator, asked what could and would be done to avoid such extensive damage and prolonged outages in future storm events (pers. comm. PSE). This article reports on work done to examine the rela- tionship between the 1998 to 2007 wind and the tree-related outage experience on PSE’s transmission system. Spe- cifically, as the regulator was concerned about electric sys- tem performance during storms, this work sought to make tree-related outage response to wind loading predictable. Utilities are charged with maintaining a reliable electric system and in some cases are financially penalized for fail- ure to meet reliability targets, which is the case for PSE. Con- sequently, it is important for utilities to understand what is within their control and what conditions will simply over- whelm their system and efforts to provide continuous ser- vice. Through this understanding, realistic system perfor- mance expectations can be advanced and communicated, creating in the process, common, shared, fact-based expectations. METHODOLOGY Two data sets were examined to assess the role of trees in PSE’s service interruption experience. One data set was the out- age statistics generated from a combination of equipment function and servicemen consisting of more than 137,000 records. When this system reports a tree-caused outage oc- curring on a transmission line or on a whole distribution circuit, PSE sends an arborist to the site to gather detailed information, creating the second data set. This informa- tion includes: the nature of the outage, either a grow-in or fall-in; tree species; tree location; and condition and type of failure. Outage data generated between 1998 and the end of 2007, for the entire system, both transmis- sion and distribution were examined. The data consist- ed of more than 34,000 tree-caused outage records and more than 6,200 arborist generated investigation records. The 265 transmission outage statistics, generated from 1998 through 2007, were sorted by county. Wind data was obtained from CompuWeather. The wind data provided maximum wind speeds for all days between January 1, 1998 and December 31, 2007 with sustained winds equal to or greater than 24 km/hr. The data is com- prised of approximately 33,000 records. Data was obtained from nine airfields (county): Bellingham International (Whatcom), Fri- day Harbor (San Juan), Everett Snohomish (Snohomish), Seattle Tacoma International (King), Tacoma Narrows (Pierce), Bremerton (Kitsap), Olympia (Thurston), Stampede Pass (King/Kittitas), and Ellensburg Bowers (Kittitas). There were three counties that either did not have airfields or did not have records for one or more years for the period sampled, in which case, wind data from the closest airfield were used. Classifying both maximum daily wind speed and tree-related service interruptions by county, the relationship be- tween wind speed and tree-related outage incidents, was examined. Graphics to represent the frequency of wind events and outages associated with wind speed were developed. The troughs in the data at 32 and 45 km/hr (Figure 1), and the saw tooth pattern or variability (Figure 2), made it ap- parent that a modeling approach was needed to smooth out the data. Exponential regression was used to de- velop algorithms to predict tree-related outage events. ©2011 International Society of Arboriculture
July 2011
| Title Name |
Pages |
Delete |
Url |
| Empty |
Search Text Block
Page #page_num
#doc_title
Hi $receivername|$receiveremail,
$sendername|$senderemail wrote these comments for you:
$message
$sendername|$senderemail would like for you to view the following digital edition.
Please click on the page below to be directed to the digital edition:
$thumbnail$pagenum
$link$pagenum
Your form submission was a success. You will be contacted by Washington Gas with follow-up information regarding your request.
This process might take longer please wait
