What We Do
National Grid Partners (NGP) serves as the eyes and ears of the business, searching for innovations to keep National Grid at the forefront of the energy industry. As the corporate investment and innovation arm of one of the world’s largest investor-owned energy companies, NGP is helping drive the transition to a more digitized, smarter, and renewable energy system. NGP does so by investing and collaborating at the intersection of emerging technology and energy.
What They Do
Cogniac aims to redefine human performance, productivity, and efficiency by maximizing the value of visual data. Cogniac combines the latest no-code / low code Artificial Intelligence (AI) models, cloud computing, and big data management to create a Visual Operations Intelligence Platform that delivers transformative enterprise machine vision solutions. In short, Cogniac’s software can read images, classify objects, spot defects or perform other visual inspection tasks much like a human can.
“Enterprises across a range of verticals are discovering they don’t need Visual AI deployed in just one or two use cases – there are hundreds of use cases,” said Quinn Curtis, CEO of Cogniac.
Why National Grid Metering and Cogniac Joined Forces
National Grid Metering (NGM) is the largest gas meter asset manager in the UK, responsible for the management and maintenance of more than 8 million gas assets.
For NGM, finding a more efficient way to address any challenges with its vast portfolio of gas assets isn’t just important — it’s increasingly critical for keeping people safe and warm.
Thanks to Cogniac’s work in AI vision inspection software, cataloguing NGM’s assets may prove more manageable than ever as customers, engineers, and meter readers can digitally submit photographs of regulators for identification. This streamlines process and allows NGM to provide greater focus and attention across the business to further deliver quality and service benefits to our customers.
Image Recognition Technology Use Case
NGM and Cogniac collaborated to train an AI model to identify different makes and models of meter regulators (which may benefit from early replacement), using a variety of photographs taken from multiple angles and in various lighting conditions. These regulators, like any asset in the field, can degrade over time, and a targeted, proactive replacement program can maximize both cost effectiveness and safety for the customer.
“Industry-leading technology is necessary because there are dozens of different regulators and each looks very similar,” said Chris Wood, National Grid’s Hydrogen and Innovation Lead. “They’re similar in color and in size. But there are differences, and there are nuances.”
Photographs were deliberately varied with different angles, different lights, and different locations, as many of the meters are located in cellars and other cramped spaces. “It’s a good way of replicating the image a consumer would capture and a way to really test what Cogniac’s software can do,” said Wood.
In proof-of-concept trials on more than 800 regulator photographs, Cogniac’s solution yielded a high success rate in identifying make and model. According to Wood, enhancing asset data will be invaluable to NGM and its customers as they move into the future.
Further trials are in discussion across National Grid Metering that would help identify and confirm completed jobs, meter readings, asset health assessments and other solutions. The ultimate goal is to provide an efficient, cost-effective approach to asset management that could become another tool in National Grid Metering’s already impressive toolkit.