Method to automate gas leak detection in battery manufacturing using data from optical gas imaging system
AssigneeGM Global Technology Operations LLC
InventorsAlok Warey, Siddhartha Gupta, Hassan Ghassemi-Armaki
Life of this patent
Estimating the term…
Abstract
An automated system to provide gas leak detection during battery manufacturing includes a battery system. A video of a gas leak occurring during a manufacturing stage of the battery system includes the gas leak as a gas vapor. A location of the gas leak is determined. A leak intensity value of the gas leak is identified to determine if the gas leak is minor and is less than or within a predetermined window or threshold permitting acceptance of the gas leak without repair, or if the gas leak requires further action including repair of the battery system.
Claims7
A method to perform automated battery gas leak detection, comprising:
- collecting a video of a vehicle battery system having multiple video frames during a manufacturing stage of the battery system;
- detecting a gas leak occurring from the battery system having the gas leak present in at least one of the video frames;
- determining a physical location of the gas leak on the battery system;
- identifying a leak intensity value of the gas leak;
- subtracting a background frame of the video not having the gas leak present from the at least one of the multiple video frames of the video having the gas leak present to obtain a frame difference having image data defining the gas leak;
- identifying multiple frame differences individually having image data defining the gas leak;
- performing a summation of the multiple frame differences;
- and
- determining a sum of total pixel values in the multiple frame differences to identify an image frame of the gas leak applying pixel thresholding.
A method to perform automated battery gas leak detection, comprising:
- collecting a video of a vehicle battery system having multiple video frames during a manufacturing stage of the battery system;
- detecting a gas leak occurring from the battery system having the gas leak present in at least one of the video frames;
- determining a physical location of the gas leak on the battery system;
- identifying a leak intensity value of the gas leak;
- subtracting a background frame of the video not having the gas leak present from the at least one of the multiple video frames of the video having the gas leak present to obtain a frame difference having image data defining the gas leak;
- performing a summation of multiple ones of the frame difference, where the summation is equal to a sum of pixel values in the multiple ones of the frame difference;
- and wherein if the sum of the pixel values is greater than one of multiple predetermined thresholds, the leak intensity value defines one of a low intensity leak, a medium intensity leak or a high intensity leak.
A method to perform automated battery gas leak detection, comprising:
- collecting a video of a vehicle battery system having multiple video frames during a manufacturing stage of the battery system;
- detecting a gas leak occurring from the battery system having the gas leak present in at least one of the video frames;
- determining a physical location of the gas leak on the battery system;
- identifying a leak intensity value of the gas leak;
- subtracting a background frame of the video not having the gas leak present from the at least one of the multiple video frames of the video having the gas leak present to obtain a frame difference having image data defining the gas leak;
- saving a calibration file having multiple values of Iixy defining coordinates in individual ones of the video frames;
- identifying the location of the gas leak as an Lxy value in an image of one of the video frames;
- computing an i value where i is equal to an argument of a minimum (argmin) multiplied by an absolute value of Lxy−I′xy;
- and
- mapping at least one of the values of I′xy to a gas leak location using the calibration file.
Dependencies are inferred from the claim text.