Home // IoTAI 2025, The Second International Conference on IoT-AI // View article

Predicting Surface Roughness in Titanium Alloy Milling Machining Through Tool Wear Images

Authors:
Hariyanto Gunawan
Chi Min Chang
Am Mufarrih
Zheng Xin Su

Keywords: YOLOv7; BiLSTM; tool wear; surface roughness

Abstract:
Abstract—This study presents a deep learning-based approach to predict surface roughness in the Computer Numeric Control (CNC) milling of Ti-6Al-4V, integrating You Only Look Once (YOLO)v7 for tool wear detection with Long Short-Term Memory/Bidirectional Long Short-Term Memory (LSTM/BiLSTM) for time-series prediction. Images of tool wear are analyzed to extract wear features, which are combined with machining parameters to forecast surface roughness. Experiments were conducted on a vertical milling machine to confirm the effectiveness of the model. YOLOv7 achieved a wear detection accuracy of 92.4%, while BiLSTM attained a prediction of 82.61%, outperforming traditional LSTM. The proposed system offers a reliable solution for intelligent tool condition monitoring and machining quality control.

Pages: 14 to 15

Copyright: Copyright (c) IARIA, 2025

Publication date: July 6, 2025

Published in: conference

ISBN: 978-1-68558-286-9

Location: Venice, Italy

Dates: from July 6, 2025 to July 10, 2025