Service Restoration in Storm Conditions

Some of the major insights of leading electric utilities regarding service restoration in storm conditions is the considerable benefit that can accrue to early mobilization of field personnel. Although these benefits must be weighed against the cost of mobilizing resources for a “false alarm” (e.g. a storm that does not hit or does less damage than that forecast), the pendulum is swinging toward ensuring that enough resources are on hand early in the storm because of the importance of restoring service to the mainline feeders as quickly as possible.

Consequently, the ability to predict storms along with their location and severity is critical to optimizing a utility’s activities around storm response.

The ability to predict storms is critical to timely service restoration

Early prediction of an incoming storm allows for the timely identification of needed resources, results in quick restoration, and supports accurate communication with customers, media and governmental officials. The use of decision tools will enhance an electric utility’s ability to:

• Forecast and assess weather
• Predict and assess damage
• Predict and track resource requirements
• Estimate restoration time

Most significant weather events are “somewhat” predictable

• Hurricanes are visible days in advance…although the exact path is uncertain
• Sustained high windscan be forecasted in advance…tornadoes are unpredictable in force and location
• Storm fronts (along with lightning, wind, rain or snow) can be forecasted…but they vary in force and location
• Ice storms are anticipated…but the exact accumulation and location vary
• Heat waves can be forecasted within the week…however, impact depends on wind, cloud cover, humidity and demand response

How to identify the  “trigger point” to initiate storm mobilization

First and foremost, significant weather events cause outages (service interruptions) and restoration times are impacted by the number of outages within a prescribed time period. The figure (below) uses historical data to illustrate the relationship between the number of daily outage cases and customer minutes of interruptions (as measured by CAIDI). In this instance, CAIDI starts to degrade significantly after 20 outage cases per day (per district) and begins to accelerate (i.e. get substantially worse) at 30 outage cases per day. Consequently, as an example, this particular utility can integrate this insight into an empirically-based decision model and start pre-mobilization activities well in advance of experiencing 30 outage cases per district within a 24-hour period. The next step is to use historical data related to the causes of outages (such as wind, heat, etc,) to provides the “trigger point” for pre-mobilization based on their root cause and measurable conditions that predict them.

Wind caused outages are exponentially related to wind speed

In analyzing wind- caused outages, a main contributor to line failures and trees/non-preventable related outages is wind. The figure (right) plotting the relationship between number of outages and effective wind speed shows that the step from 35 to 40 miles per hour causes more outages (and damage) than the step from 30 to 35 miles per hour. These thresholds vary by region (and are obviously affected by actual system configuration and topography), but in plotting effective wind speed, the variables related to seasonal effects and duration of the wind events are normalized.

Lightning strokes drive lightning outages-a linear relationship

In plotting lightning outages per day against the number of lightning strokes per day there appears to be a consistent slope, but the number of strokes per day driving that slope vary with equipment density.

Excessive heat over an extended period of time causes outages

Aside from  equipment aging related issues, a leading contributor to equipment failure caused outages is excessive heat. Again, historical data can be used to establish precise points on a utility by utility basis, but typically temperatures in the 90’s for three days (or more) in a row, particularly with high humidity leads to an exponential increase in heat-related equipment failures.

Recognizing that there are a number of proactive measures that can be taken to mitigate the impacts of these weather-related variables on reliability, the point here is that any specific point in time the system (as it is then configured) will have limits regarding its ability to withstand these phenomena. And, to the extent that the inflection points are known (and adjusted as these proactive measures are implemented), the electric utilities can glean valuable insight regarding the optimal deployment of resources in anticipation of system outages. As previously stated, combined with the intuition and knowledge of its experienced professionals, the conundrum of early mobilization and potential “false alarms” can be somewhat abated.