Back to results
US 10,947,849 B2Granted
Systems and methods for recontouring engine components
AssigneeGeneral Electric Company
InventorsJohn Karigiannis, Jessen Compagnat
Life of this patent
Estimating the term…
Abstract
Systems and methods that include and/or leverage a cluster of machine-learned models to recontour components of gas turbine engines are provided. In one exemplary aspect, the systems and methods leverage a cluster of machine-learned models to predict repair machining offsets for certain sections of the component that can be used to adjust or set a material removal tool path.
Claims19
A method for recontouring a component defining one or more sections, the method comprising:
- obtaining, by one or more computing devices, a data set comprised of a plurality of subsets each comprised of one or more parameters indicative of a condition of one of the sections of the component and one or more parameters indicative of a condition of a recontouring system;
- inputting, by the one or more computing devices, the data set into a cluster of machine-learned models, each subset being input into a respective machine-learned model;
- determining, by the one or more computing devices, a machine offset for each section of the component based at least in part on the one or more parameters indicative of the condition of the section and based at least in part on the one or more parameters indicative of the condition of the recontouring system;
- obtaining, by the one or more computing devices, the machine offset for each section of the component as outputs of their respective machine-learned models;
- adjusting, by the one or more computing devices, a material removal tool path of the recontouring system based at least in part on the machine offsets;
- and
- machining the component utilizing the material removal tool path, and
- wherein the one or more parameters indicative of the condition of the recontouring system include a measured step, wherein the measured step is measured during a post-inspection process of a first component defining one or more first sections, the first component being machined prior to the component by the recontouring system and defining a regenerated portion and a non-regenerated portion after being machined, the measured step being measured along the first component for each of the first sections as a distance between an edge of the non-regenerated portion to an edge of the regenerated portion.
5 more elements
A method for recontouring a component of a gas turbine engine using a recontouring system, the component defining a plurality of sections, the method comprising:
- obtaining, by one or more computing devices, a data set comprised of a plurality of subsets, each subset comprised of one or more parameters indicative of a condition of one of the sections of the component and a condition of the recontouring system;
- inputting, by the one or more computing devices, the data set into a machine-learned model comprised of a plurality of neural networks, each subset being input into a respective neural network;
- determining, by the one or more computing devices, a machine offset for each section of the component based at least in part on the one or more parameters indicative of the condition of the section and the condition of the recontouring system;
- obtaining, by the one or more computing devices, the machine offset for each section of the component as outputs of their respective neural networks;
- adjusting, by the one or more computing devices, a material removal tool path of the recontouring system based at least in part on the machine offsets;
- and
- machining the component utilizing the material removal tool path, and
- wherein the parameter indicative of the condition of the recontouring system is a measured step, wherein the measured step is measured during a post-inspection process of a first component defining one or more first sections, the first component being machined prior to the component by the recontouring system and defining a regenerated portion and a non-regenerated portion after being machined, the measured step being measured along the first component for each of the first sections as a distance between an edge of the non-regenerated portion to an edge of the regenerated portion.
5 more elements
A method for recontouring a blade of a gas turbine engine using a recontouring system, the blade defining a plurality of sections, the method comprising:
- obtaining, by one or more computing devices, a data set comprised of a plurality of subsets, each subset corresponding to one of the plurality of sections of the blade and comprising one or more parameters indicative of a condition of the corresponding section of the blade and a condition of the recontouring system;
- inputting, by the one or more computing devices, the data set into a machine-learned model comprised of a plurality of neural networks, each subset being input into a respective neural network;
- determining, by the one or more computing devices, a machine offset for each section of the blade based at least in part on the one or more parameters indicative of the condition of the section and the condition of the recontouring system;
- obtaining, by the one or more computing devices, the machine offset for each section of the blade as outputs of their respective neural networks;
- adjusting, by the one or more computing devices, a material removal tool path of the recontouring system based at least in part on the machine offsets;
- and
- machining the blade utilizing the material removal tool path, and
- wherein the parameter indicative of the condition of the recontouring system is a measured step, wherein the measured step is measured during a post-inspection process of a first component defining one or more first sections, the first component being machined prior to the component by the recontouring system and defining a regenerated portion and a non-regenerated portion after being machined, the measured step being measured along the first component for each of the first sections as a distance between an edge of the non-regenerated portion to an edge of the regenerated portion.
5 more elements
Dependencies are inferred from the claim text.