GoogleGranted patentAI & softwareInterest 6/10

Google patents method to predict power grid failures from images

Patent drawing
Patent drawing from the official filing.

This patent describes a system that uses computer vision to monitor electrical grid components. It compares images of a part taken at different times to spot defects and uses machine learning to estimate the chance that the part will fail.

Why it mattersThis could help utility companies fix power lines before they break, potentially reducing the frequency of unexpected blackouts.

In the filing’s own words

Methods, systems, and apparatus, including medium-encoded computer program products, for predicting electrical component failure. A first sensor measurement of a component of an electrical grid taken at a first time can be obtained. A second sensor measurement of the component taken at a second time can be identified, and the second time can be after the first time. An input, which can include the first sensor measurement and the second sensor measurement, can be processed using a machine learning model that is configured to generate, based on one or more changes in one or more characteristics of the component as depicted in the second sensor measurement compared to the first sensor measurement, a prediction representative of a likelihood that the component will experience a type of failure during a time interval. Data indicating the prediction can be provided for presentation by a display.

What exactly they claim (claim 1)

1. An electrical grid asset failure prediction method comprising: obtaining a first image that depicts a component of an electrical grid within a portion of the first image at a first time; identifying a set of second images that each depict the same component of the electrical grid within the images and each image represents the component of the electrical grid at a respective time after the first time; processing, in accordance with trained values of a set of parameters of a first machine learning model, the first image and the set of images as first input to the first machine learning model that is configured to detect electrical component defects within images and to generate image encodings that indicate a presence and a type of any detected defects, the first machine learning model output including a set of image encodings indicating a particular defect in each of the first image and the second images; processing, in accordance with trained values of a set of parameters of a second machine learning model, the set of image encodings as second input to the second machine learning model that is configured to determine a likelihood that the component of the electrical grid will fail based on a type of the particular defect and a rate of change of the particular defect as identified in each of the first image and the second images; and providing, for presentation by a display, the likelihood that the component of the electric grid will fail.

Facts

Company
Google (X Development LLC)
Official title
Predicting electrical component failure
Patent number
US12803278B2
Application number
18331765
Filed
Jun 8, 2023
Granted
Nov 17, 2026
Inventors
7
Claims
14

Classified under

H02J 3/001 Electricity & electronicsG01N 29/14 Physics & computingG06T 7/0004 Physics & computingG06T 2207/10048 Physics & computingG06T 2207/20081 Physics & computing

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.