UI and Eversource Report Long-term Improvements in Service Reliability

Connecticut’s for-profit energy providers improved their day-to-day reliability in 2019, according to an annual report on the reliability of the state’s electric system compiled by its utility regulator, PURA.

With the exception of major storms, United Illuminating and Eversource reported shorter and less frequent average outages compared to the previous four years, according to a report approved on Wednesday morning. 

PURA is required by state statute to submit the annual reliability report to the General Assembly.

The report includes two metrics: one for frequency and one for duration of service interruptions. The System Average Interruption Frequency Index (SAIFI) measures how many power outages the average customer had to endure, measured by the number of customers experiencing an outage divided by the total number of customers. 

The System Average Interruption Duration Index (SAIDI) measures how long the average customer went without power, and is calculated by multiplying the number of minutes an outage lasted by the number of customers affected, and dividing by the total number of customers in the system.

For Eversource, the average customer in 2019 experienced 0.573 power outages and 63.3 minutes without power. From 2015-2018, the average customer had experienced 0.66 power outages totaling 72 minutes without power each year. 

For United Illuminating, the average customer in 2019 experienced 0.5 outages and 38 minutes without power. From 2015-2018, the average customer experienced the same number of outages and 43 minutes without power each year.

According to the report, each measure was a significant improvement over the four years before the electric system was deregulated in 1998, when the average Connecticut Light & Power customer experienced 1.22 outages and 132 minutes without power a year, and the average UI customer experienced 0.77 outages and 52 minutes without power a year from 1995-1998.

PURA spokesperson Taren O’Connor said that the information helps their understanding of how service reliability has changed over time.

“It is relevant to, and important context for, understanding the utilities’ vegetation management and other reliability-related programs, but is not the other information that PURA looks at when assessing those programs,” O’Connor explained in an email to CT Examiner.

What caused the outages?

Counting the number of disruptions, United Illuminating customers experienced 34 percent more outages in 2019 than the previous four years. Almost half of that increase can be attributed to additional outages caused by equipment failure, of which there were 1,233 in 2019 – 74 percent more than the four-year average of 711.

Outages caused by trees increased by 30 percent for United Illuminating, with trees causing 708 outages in 2019, compared to an average of 545 the four years prior. Outages caused by birds and animals were the only category to decrease from the four year average, down 65 percent from 368.25 outages a year to 130 in 2019.

While UI experienced more outages, Eversource reported 13.4 percent fewer service disruptions in 2019 compared to the average of the prior four years. Planned outages increased 31 percent to 1,504 in 2019, and outages caused by lightning increased 13.8 percent to 554, but outages for Eversource declined in every other category.

Eversource saw a sharp decrease in the number of outages caused by equipment failures (down 25.4 percent from the four year average to 2,364 outages in 2019) and those caused by animals and birds (down 35 percent to 1,887).

What about big storms?

By state statute, PURA does not include major storms, planned outages or outages caused by customers in their official measures, though they do include measurements calculated to include major storms in the report. 

O’Connor said reliability, measured in this report, refers to how the system performs on “blue sky days” without major storms, while resilience refers to the system’s ability to withstand and recover from bad weather.

Including major storms, Eversource recorded outage durations lower than the four year average (249.2 minutes compared to 360 minutes), but a slightly higher frequency of outages (1 in 2019 compared to 0.99 from 2015-2018). 

UI reported greater service disruptions both in frequency and duration when including major storms, with customers experiencing 0.85 outages and 205 minutes without power, compared to 0.67 outages and 112 minutes without power a year over the prior four years. 

Eversource had eight “major” storm events in 2019.

The largest by far were wind storms on Oct. 16 and Oct. 31 that caused the loss of 1.1 million and 1.65 million “customer-hours” of power, respectively. “Customer-hours” are the total number of hours of power loss across all customers. In total, Eversource lost 7.69 million customer hours to major storm events in 2019.

But the number of Eversource outages caused by major storms declined almost 5 percent in 2019 from the average of the prior four years, with storms causing 6,916 outages for Eversource customers in 2019 compared to the four-year average of 7,260 outages a year.

United Illuminating experienced 10 “major” storms in its service area in 2019, knocking out power for a total of 494,779 “meter-hours” – just like “customer hours,” but counting meters. By far the largest was a Jan. 20 winter storm that caused the loss of 205,555 meter-hours of service.

The number of outages caused by major storms in the UI service area increased by almost 55 percent to 1,205 outages in 2019, compared to the four-year average of 780.25 outages a year.

Performance-based ratemaking?

“Performance-based ratemaking” was a centerpiece of the bill the Connecticut General Assembly passed in October, aimed at increasing accountability for the electric distribution companies Eversource and UI. 

The new law allows PURA latitude to set up the metrics — but calls for the new ratemaking system in general to allow PURA to raise or lower rates depending on whether the companies meet customer needs, based on a range of metrics including reliability, cost and emergency response.

O’Connor, said that, theoretically, the reliability data in this report could help set reliability-related metrics for performance-based rates.

“PURA has not yet formulated a position as to whether reliability data and metrics should be included in performance-based rates,” said O’Connor.

State Sen. Norm Needleman, D-Essex, co-chair of the General Assembly’s Energy and Technology Committee, said reliability metrics should be a major factor in performance-based rates.

“The standards should include reliability metrics, they should include communication, they should include a whole bunch of factors that were not necessarily part of the equation,” Needleman said.

Needleman said Eversource has gotten “significantly better” at both day-to-day reliability and responding to more localized power outages, like outages caused by the tornado in Branford in August.

“I think they did a great job there – and tornadoes are the worst, you have to rebuild the whole system, and they did a good job,” Needleman said. “When we had [Tropical Storm Isaias] it just showed their weakness.”

A key weakness for Eversource was a lack of communication — and the company has recognized that — Needleman  said.

“You could be doing a stellar job, but if you’re not communicating with the customers, with the town leaders, you’re just not doing a good job,” he said.

Latest from Brendan Crowley