Many electric utilities are deploying advanced voltage optimization (VO) technology for energy conservation: VO can save up to 4% of the energy supplied to a grid. Deployment of VO technology can illuminate existing grid issues, and addressing those areas can help utilities to further improve reliability, performance, safety and customer service. 

“Historically, most distribution voltage issues have not been obvious to utilities or their customers,” explained Todd Headlee, director of DVI, a leading provider of VO technology. “Utilities must maintain voltage within a specified range, but the U.S. market does not have firm penalties for straying outside that range. Elsewhere, such as in Australia, distribution system operators are actively fined for going outside the voltage band.”

Consequently, it’s unusual that a U.S. utility will consistently monitor voltage levels along feeders, after power leaves the substation. Occasionally a customer may complain about voltage that’s too low or too high — but it’s far more likely that customers will assume that their malfunctioning devices and appliances are defective, before suspecting problems with power grid voltage. 

“Once utilities implement VO, they’ll start to see where voltage irregularities exist across their distribution grid,” Headlee said. “Then they can address these proactively to prevent problems for customers and for the grid.”

In January 2018, Versant Power (formerly Emera Maine) began testing DVI’s VO solution at a typical four-circuit substation: a suburban location that includes some long feeders as well as some commercial customers. Most of the meters in the test area were smart meters. However, Versant soon discovered that not all smart meters are created equal: many of their installed meter base were older AMI (advanced metering infrastructure) technology.

“We learned that 90% of the 6,000 smart meters in the test area did not provide sufficient resolution on voltage readings to support the level of voltage reduction we’d planned to try,” said Debbie Manning, senior transmission planning engineer for Versant. “We had to adapt our pilot plan to include mixed meter types.”

To test VO, Versant had to identify a “bellwether set” of meters: the lowest-voltage meters on each circuit in the test area. Basing voltagereduction activity on this set ensures that no meter will experience undervoltage. Normally, DVI’s software automatically rechecks the network periodically and defines a new bellwether set based on changing grid conditions. However, since Versant had so few suitable smart meters in the initial test, DVI guided the utility to define a fixed bellwether set, based on hourly voltage readings.

This process yielded a useful surprise. “We expected that our lowest-voltage meters would be located at the extremities of circuits. But actually, that’s not where we saw the voltage issues,” Manning said. “Instead, the main causes of our voltage drops were at distribution transformers on a circuit, or at long secondary feeds to some homes.”

Versant realized that it had to do more analysis on the AMI data it had gathered. “Voltage optimization technology helped us see systemic issues that we hadn’t noticed before,” Manning said. “We had been reading our smart meters for energy usage, but not for analyzing voltage data. Now we can target repairs and upgrades with voltage levels in mind.”

During Versant’s initial VO test, the utility also discovered issues with the cellular data communications used to transmit command signals to line regulators. In rural terrain that is often hilly and forested, with long distances between cell towers, signal transmission and reception proved difficult — which diminished energy savings from the pilot somewhat. As the utility rolls out VO across its network, more reliable communications will improve VO performance. The energy savings benefit to ratepayers actually makes the business case for these smart grid investments.

“During our one-year test of conservation voltage reduction, we averaged energy savings of 1.5%. But we can probably get more now that we know what to address,” Manning said. “You can really learn a lot about your system from an initial test of voltage optimization.”

Versant’s VO test also helped the utility to identify a new opportunity to accommodate the state’s rapidly-growing supply of solar power — which in relative terms is expected to exceed proportionate rates of solar penetration in California. Versant is considering importing data from smart inverters at solar generation into its VO control software to help flatten distribution voltage profiles across each circuit. “This would allow solar providers to come online safely to Versant’s grid as we monitor and dynamically control the voltage profile on those circuits,” Manning said.

Some of the voltage issues clarified by VO are easy to remedy. For instance, Headlee said, a VO pilot test in North Carolina revealed that on one circuit, voltage levels never got higher than 108 volts — while ANSI standards require a voltage range of 114v to 126v. “Nobody was complaining. Homeowner devices didn’t fail immediately, and nobody blamed their device issues on the utility,” he said. “The fix was very simple. A transformer had been configured wrong. It took 10 minutes to fix.”

“There are many ways, large and small, that VO can help utilities enhance operations,” Headlee said. “Without it, they may never see all the voltage issues that have been chipping away at grid performance. That can be death by 1,000 paper cuts.”