NYPA project uses AI to better understand performance of underwater cable

The New York Power Authority (NYPA) is launching a demonstration project that will use artificial intelligence (AI) as part of the potential upgrade of underwater cable that transports power from Westchester to Long Island.

Working with Eneryield, a Swedish company that provides machine learning algorithms for intelligent energy analytics and electricity flows, NYPA plans to use the technology to identify possible solutions, to detect faults, and help strengthen and upgrade the Long Island Sound Cable, which is being evaluated for long-term repairs.

The project will focus on NYPA’s 23-mile, 693 megawatt (MW) Y49 transmission cable. Historical data will be used from various sources and artificial intelligence/machine learning techniques will be applied to identify small anomalies, deviations and patterns to predict larger imminent disturbances or faults.

The aim of the project is to determine whether the technology can help predict developing problems or incipient failure of buried and underwater cables and improve on unique correlations and data characteristics that can be measured in more conventional analysis techniques.

The Long Island Sound Cable has had faults that have contributed to intermittent outages over the past year. NYPA is working with its local partners, the Long Island Power Authority and its service provider, PSEG Long Island, to implement a long-term strategy for the cable’s future reliability and resiliency.

Potential solutions include replacing segments of the span and possibly expanding the line’s capacity to prepare for an influx of green energy sources. The results of the demonstration project will help inform next steps for the line’s upgrade.

“This is an opportunity to take new technologies that have shown promise in development and put them to the test with real-time data and an active power system,” Alan Ettlinger, senior director of research, technology development and innovation for NYPA, said in a statement. “The use of artificial intelligence in infrastructure inspections can help increase reliability and safety, recognize malfunctioning equipment and identify problems that need repair, therefore mitigating outages for customers.”

The Electric Power Research Institute’s (EPRI) Incubatenergy Labs program recently selected Eneryield as one of 20 startup companies that will conduct accelerated demonstrations of their technologies with utilities and EPRI as part of Incubatenergy Labs’ 2021 cohort.

The 20 startups selected through Incubatenergy program the will spend 16 weeks working with EPRI and electric power utilities around the country on demonstration technology projects intended to accelerate decarbonization, electrification, grid modernization and other electric power industry innovation imperatives. The results will be presented during EPRI’s Incubatenergy Labs Demo Days in mid-October.

EPRI is one of several entities that are exploring the use of artificial intelligence in the electric power industry.

The use of AI to monitor transmission cable performance is only one of several uses NYPA is studying for the technology. In May, NYPA selected C3 IoT to provide a software platform to help the it and the state implement and meet its energy efficiency targets.

In June, the Tennessee Valley Authority teamed up with Oak Ridge National Laboratory and the University of Tennessee System to study the use of AI in a variety of applications, including cybersecurity, digital currency, 5G broadband cellular technology, and other innovations.

Also in June, independent power producer Vistra said it plans to use AI at its Moss Landing energy storage facility in California to help it better predict wholesale power market prices.