Automatic Borescope Damage Assessments for Gas Turbine Blades via Deep Learning

TL;DR

This paper introduces an automatic deep learning-based workflow for detecting damage on gas turbine blades from borescope videos, aiming to improve inspection speed and accuracy while reducing human bias.

Contribution

It presents a novel automated damage assessment method using deep learning, specifically tailored for borescope video analysis of turbine blades.

Findings

Effective damage detection demonstrated on two borescope videos.

Damage statistics can be generated for individual blades within a blade row.

Workflow reduces manual effort and potential human bias in inspections.

Abstract

To maximise fuel economy, bladed components in aero-engines operate close to material limits. The severe operating environment leads to in-service damage on compressor and turbine blades, having a profound and immediate impact on the performance of the engine. Current methods of blade visual inspection are mainly based on borescope imaging. During these inspections, the sentencing of components under inspection requires significant manual effort, with a lack of systematic approaches to avoid human biases. To perform fast and accurate sentencing, we propose an automatic workflow based on deep learning for detecting damage present on rotor blades using borescope videos. Building upon state-of-the-art methods from computer vision, we show that damage statistics can be presented for each blade in a blade row separately, and demonstrate the workflow on two borescope videos.