The Holistic Approach: Proactive Voltage Monitoring Drives Long-Term Payoff
Smart meters are providing data that has never been available beyond the substation. One of the best measures of the quality of a utility's product is voltage, which is measured at the point of sale for every utility customer.
By Jeff McCracken
Smart meters are providing data that has never been available beyond the substation. One of the best measures of the quality of a utility’s product is voltage, which is measured at the point of sale for every utility customer.
Voltage monitoring allows the utility to understand voltage levels throughout its distribution network and its impact on customers being served. Voltage monitoring via smart meters can also detect high or low-voltage problems anywhere on the distribution network.
By proactively resolving conditions before they become serious issues, a utility can increase customer satisfaction and prevent equipment damage. Ongoing monitoring of the system enables analysts to develop holistic solutions to voltage problems rather than simply react to the most recent customer complaint or problem. Following are just a few of the many advantageous outcomes voltage data provides:
- Identify high or low voltage trends on circuits
- Resolve voltage-related customer complaints
- Troubleshoot high or low voltage conditions
- Improve Volt/VAR optimization and conservation voltage reduction (CVR)
- Identify transformer winding failures
- Identify meter phase
- Identify and validate meter-to-transformer connectivity
Analysis of voltage data is key to achieving these benefits. With meter phase detection and transformer-to-meter connectivity analysis and validation, utilities can realize additional benefits such as improving reliability with more accurate outage monitoring as well as improving asset management programs.
Meter Phase Detection
Meter phase identification accurately determines which of the three phases (A, B or C) is connected to each individual meter and corresponding service point. Automated meter phase identification has recently become a practical option for utilities using advanced metering infrastructure (AMI) data. Historically, identifying the phase connection of each distribution service point and transformer has been a costly and time-consuming process involving visual inspection or signal injection. Utilizing AMI measurements to accomplish this task in an automated manner enables analysis on a regular basis, delivering more accurate and up-to-date connectivity information to support the utility’s distribution planners and engineers.
Using sophisticated machine-learning techniques, Itron’s scientists have developed and patented algorithms to accurately determine phase connectivity and also identify which meters are connected to which distribution transformers. These methods use interval voltage measurements as the basis of their analysis. The benefits of having this information includes energy balancing for loss evaluation and phase balancing. With energy balancing, the total energy at the feeder head is compared to aggregated measurements at all locations downstream on the feeder. The difference is a measurement of equipment losses, line losses and theft. Phase balancing matches the load on each phase of a three-phase distribution circuit, which is critical to safe and efficient energy delivery.
Transformer-to-Meter Connectivity Analysis and Validation
Soon after new service installation, the utility may start with a fairly accurate meter-to-transformer association. Over time, intermittent unscheduled power outages may require utilities to rewire specific meters to different transformers. Not all changes find their way into utility databases. As time progresses, actual meter-to-transformer association tends to drift from what is encoded in utility records. In some cases, services may have been installed many years ago prior to the utility adopting a formal process that tracks the meter-to-transformer association. Consequences of connectivity errors are faulty transformer load management, inaccurate outage locations and faulty CAIDI and SAIDI calculations due to mis-associated meters or missing transformers that may be shown in a GIS map, but are not physically installed in the field.
Itron has developed a new patent-pending algorithm that determines meter-to-transformer association from five-minute voltage data collected over a seven to 30-day period. Applying state-of-the-art machine learning techniques, meters are grouped by most likely transformer association and compared to utility records for validation or correction or both. This algorithm correlates voltage changes over time between individual meters and defines the meter “affinity.” The higher the affinity the more likely these meters are connected to the same distribution transformer.
There are several benefits that result from an accurate meter-to-transformer model, including the ability to enable asset management, outage management, historic outage analysis and load balancing as well as improve the accuracy of transformer load analysis programs.
Smart meters provide several outage-related critical capabilities, including real-time power monitoring and outage alarms, two-way communication for power-on validation and acquisition of historical outage data from each individual customer, with accuracy down to the second. This data, combined with an accurate connectivity model, allows the utility to better understand the extent of the outage and the associated restoration effort. This capability also allows the utility to understand precisely where momentary and sustained outages are occurring in the system, providing them with the focus to more effectively improve their reliability efforts. These capabilities can have significant improvements on a utility’s overall reliability index ratings.
With the adoption of advanced metering infrastructure, utilities now have secondary voltage-sensing capabilities at each customer service point, providing visibility into grid operations that were historically unavailable. Some smart meters support voltage profiles at several interval lengths for average, maximum and minimum voltage readings in addition to exceptions for high and low voltage occurrences. This voltage monitoring information can be used to identify high or low voltage conditions, validate customer complaints, support VVO/CVR programs, and identify how meters are connected to transformers and to which phase the meter is connected.
Voltage bellwether meters can be dynamically assigned to monitor distribution assets and feeder sections. It is best for a utility to implement a voltage configuration and collection strategy that will allow the utility to achieve these benefits with its existing meter and network infrastructure without impacting more conventional billing operations.
Implementing a holistic voltage monitoring program can result in significant benefits to a utility in several areas, including operational costs, energy efficiency, system reliability, extended capital asset life and increased customer satisfaction. These drivers have always been important to utilities; however, with the introduction of distributed generation on the grid, utilities must now manage a more dynamic two-way infrastructure with significant load increases as electric vehicles and other distributed generation resources become more prevalent.
By providing distribution planners and engineers with an awareness of changing voltage conditions, utilities can continue to provide the high quality level of service expected by customers and regulators.
Jeff McCracken is a senior product line manager at Itron Analytics. He provides global leadership for Itron Analytics’ solutions that incorporate big data technology, industry expertise and advanced algorithms to deliver business intelligence and predictive analytics to Itron’s utility customers. Jeff oversees the discovery and validation of market opportunity and market use cases.