Active seismic faults in the Pacific Northwest area have encouraged electric utilities in the region to deeply contemplate and proactively intervene to support grid resilience. To further this effort this research introduces Monte Carlo (MC)-based power system modeling as a means to inform the Performance Based Earthquake Engineering method and simulates 100,000 sample earthquakes of a 6.8 magnitude (M6.8) Portland Hills Fault (PHF) scenario in the Portland General Electric (PGE) service territory as a proof of concept. This paper also proposes the resilience metric Seismic Load Recovery Factor (SLRF) to quantify the recovery of a downed power system and thus can be used to quantify earthquake economic risk. Using MC results, the SLRF was evaluated to be 19.7 h and the expected economic consequence cost of a M6.8 PHF event was found to be $180 million with an annualized risk of $90,000 given the event’s 1 in 2000 year probability of occurrence. The MC results also identified the eight most consequential substations in the PGE system—i.e., those that contributed to maximum load loss. This paper concludes that retrofitting these substations reduced the expected consequence cost of a M6.8 PHF event to $117 million.

No abstract

No abstract

The real time monitoring for the Olive 8 project in Seattle is described. This project involved construction of a 39 story structure with 5 levels of below grade parking. The 76 ft deep excavation next to an adjacent building required a hybrid support system consisting of large soldier piles with tightly spaced soil nails in the upper portion and steeply inclined tieback anchors in the lower portion. Because the City of Seattle limited lateral movements to 1 inch, engineers implemented a detailed monitoring system. Conventional inclinometers and optical survey points were supplemented by real time monitoring consisting of a web camera and an automated total station. This paper summarizes the philosophy behind the real-time instrumentation and shows how the total station data complemented the conventional inclinometer data. Data reduction techniques are described and performance data are presented. Finite element simulations of the excavation process showed the critical observation was the cantilever movements at the top of the support walls. Survey prisms were mounted at three such locations along the west wall to provide real time performance data. A project web site allowed all interested parties access to the data. The paper illustrates how real time monitoring and careful design can be used to great effect.

No abstract