Whether it's during a coastal hurricane or blue skies, knowing about outages before customers call can dictate whether utilities achieve top performance in customer satisfaction, restoration speed and restoration cost.
Knowing about outages before customers call is one of the key desired outcomes of utility outage and restoration management (ORM) efforts. Feeding outage notifications from smart meters to the outage management system (OMS) shows promise, but false positives can cause unintended headaches for utilities. Utilities such as BC Hydro are leveraging these technologies for outage and restoration notification, but in a much smarter way.
Intelligent Filtering for AMI-OMS Integration
In 2011, BC Hydro began implementation of its Smart Metering & Infrastructure (SMI) Program that eventually would lead to a full deployment of more than 1.8 million smart meters installed across BC Hydro's urban and rural service territories.
This program was designed to provide benefits including the ability to notify the utility when a customer loses power and when power is restored.
Like many utilities that implement advanced metering infrastructure (AMI), BC Hydro wanted to use the outage messages from smart meters to:
- Know when customers lose power;
- Reduce the time it takes to learn about outages; and
- Help the utility understand the scope of outages more quickly without having to wait until customers call.
Feeding power outage notifications (PONs) from smart meters into the OMS provides a better picture of the outage scope more quickly.
Unfortunately, a simple integration between the meter head-end system and the OMS can lead to false positive PONs, which can obscure real outage information. When this occurs, operators can't make good use of new outage information and the utility cannot achieve the operational efficiency it desires. A more intelligent filter often is required to provide actionable data points.
Filtering Outage Messages
The purpose of the filter is to collect messages from the AMI head-end system, filter those messages and then pass on validated outages to the OMS. Filtering occurs in three stages:
- Stage one: eliminates duplicate power-off notifications;
- Stage two: filters momentary interruptions, such as those caused by reclosers; and
- Stage three: filters messages from "cold-style" meters.
Cold-style meters include a switch between the secondary and the meter that can be operated by the building owner. If the building owner opens the cold-style switch, the meter loses power and sends a power-off notification. In such a case, the PON does not indicate a problem with the utility supply and should be filtered.
In addition to enhanced filtering logic to eliminate unwanted outage messages, the solution requires a second piece called the pinger.
The pinger checks the operational status of meters upon request. This helps customer service representatives check power-on or power-off status of a meter to determine if the outage is on the utility or customer side.
This capability can help customer service diagnose a customer problem before rolling a truck to the customer's location (see Figure 1).
Implementing Intelligent Filtering at BC Hydro
In the case of BC Hydro, the first step was to figure out where the filtering should be done based on the configuration of the utility's technology.
Filtering could be done in three places. After testing several options, the utility worked with and leveraged Bridge Energy Group's related experience to develop an intelligent filtering algorithm as middleware that would filter out false positives.
BC Hydro uses the TIBCO enterprise service bus for integration between its information technology systems.
The team leveraged the business events capability in the TIBCO ESB to build the intelligent filtering algorithm. The intelligent filter then incorporated the capabilities of all three filtering points (see Figure 2).
The solution was completed and implemented in 2014. Initial results show that the filter is eliminating 70 to 80 percent of unwanted outage messages coming from the head-end system.
The remaining outage messages can be passed on to the OMS.
At this time the outage messages are being queued in the OMS and manually accepted by operators.
BC Hydro plans to enable the automatic generation of outages in the OMS using the outage messages from the filter.
While the solution was implemented using the typical ESB and a power-on OMS, the filtering logic and pinger principles used here can be implemented in any environment.
All that is required is a platform in which to build the filtering logic.
A Vision for ORM
No matter whether a utility is focused on smart metering and AMI as at BC Hydro or implementing advanced distribution management systems (DMS), there are five key desired outcomes that most utilities are looking to achieve when they think about ORM.
Know about outages before customers call.This is addressed most often with smart metering and AMI, where these technologies are used for outage and restoration notification; however, there will continue to be ways to apply other sensors and communications technologies to identify power outages more quickly and accurately.
Diagnose and isolate outages to affect fewer customers. An OMS is a primary tool for this, along with supervisory control and data acquisition. Increasingly, advanced DMS are being implemented along with distribution automation in the field. Capabilities such as fault location isolation and service restoration (FLISR) promises to support a dramatic improvement in this area.
Dispatch the right crews to the right location with the right equipment to make the repair.Coordinating OMS with technologies such as automatic vehicle location, work management systems and other mobile technologies are key leverage points here.
Provide accurate estimated time to restoration (ETRs) for all outages. Accurate ETRs are one of several types of information that can provide value to customers during a power outage. Power quality and reliability are key drivers of overall customer satisfaction, but the weakest area of performance is keeping customers informed about outages, according to J.D. Power.
Regardless where a utility falls on the continuum, outage and restoration management is complex. From the time the power goes out to the time power is restored, numerous business processes and enabling technologies are involved. Improving operational performance starts with applying technology at key points along the end-to-end outage management process (see Figure 4).
Targeting Measurable Benefits
Each outcome produces measurable benefits for utilities and consequently their customers (see Figure 5). Achieving business case benefits depends on how well the utility can isolate the mechanisms that technology and process changes support and how well the change-benefit-can be valued or monetized. For example, reducing the number of truck rolls or optimizing resources can be translated directly to a monetized business case benefit. These are relatively easier to determine. By contrast, the strength of the business case for improving reliability statistics or providing better outage information depends on how the utility and regulator can reach agreement on recognizing and paying for improved performance.
Utilities must get specific about what types of improvements they want to make and then track performance with metrics that align with those improvements. For utilities such as BC Hydro, eliminating 70 to 80 percent of unwanted outage messages yields shorter outages, lower restoration costs and higher customer satisfaction. That's a meaningful return for the utility and its customers.