Back to results
US 11,446,763 B2Granted
Nozzle state or type identification in a metal machining apparatus
AssigneeBystronic Laser AG
InventorsAndreas Luedi, Christoph Fahmi, Kevin Messer
Life of this patent
Estimating the term…
Abstract
The present disclosure relates to a metal machining apparatus having a gas nozzle for generating a gas jet. The apparatus also has a nozzle exit opening on one end on the outside; an electronic camera for acquiring a digital image of the end of the gas nozzle with the nozzle exit opening. The apparatus also includes a pattern recognition module for mapping the digital image to at least one nozzle pattern from the group of nozzle state and/or nozzle type.
Claims15
A machining apparatus for metal, comprising:
- a gas nozzle configured for generating a gas jet, the gas nozzle having a nozzle exit opening at one end on an outside of the gas nozzle;
- an electronic camera configured for acquiring a digital image of the end of the gas nozzle with the nozzle exit opening;
- a pattern recognition module for mapping the digital image onto at least one nozzle pattern from at least one of the nozzle state group and nozzle type;
- and
- wherein the pattern recognition module comprises a trained neural network or a deep learning algorithm and means for detecting the nozzle state and the nozzle type, and
- wherein the neural network is initially trained and taught on the basis of existing image data of used and unused gas nozzles, which have previously been assessed and classified by a cutting expert and wherein the weights between individual neurons and weights of the neurons themselves are set, so that after successful training, the algorithm can automatically detect the nozzle state, in particular, a degree of wear according to the trained criteria and degrees as well as the nozzle type.
3 more elements
A method for detecting a state of wear of a gas nozzle, the method comprising the steps of:
- detecting a digital image of an end of a gas nozzle having a nozzle exit opening;
- mapping the digital image to at least one nozzle pattern from at least one of the nozzle state group and nozzle type with a pattern recognition module;
- wherein the pattern recognition module comprises a trained neural network or a deep learning algorithm;
- and
- detecting the nozzle state and the nozzle type and wherein the neural network is initially trained and taught on a basis of existing image data of used and unused gas nozzles, which have previously been assessed and classified by a cutting expert and wherein the weights between individual neurons and weights of the neurons themselves are set, so that after successful training, the algorithm can automatically detect at least one of the nozzle state, a degree of wear according to the trained criteria and degrees, and the nozzle type.
2 more elements
Dependencies are inferred from the claim text.