Tool Wear Prediction Under Varying Cutting Conditions: A Few-Shot Warm-Start Framework Based on Model-Agnostic Meta-Learning
Machines · 23.04.2026 · 14 (5) · 471
DOI: 10.3390/machines14050471 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machines, online erschienen am 23.04.2026. Verfasst von Ju Zhou und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shenzhen Polytechnic University, National Key Research and Development Program of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
In high-value precision machining, existing tool wear monitoring models often suffer from two major limitations: poor generalization under varying cutting conditions and heavy reliance on large amounts of labeled data for new operating scenarios. These limitations hinder the practical deployment of intelligent monitoring systems. To address these challenges, this paper proposes a few-shot warm-start framework based on model-agnostic meta-learning. The method consists of two stages. First, meta-training is performed on historical machining data to learn a task-sensitive parameter initialization that enables rapid adaptation. Second, under a new operating condition, the few-shot warm-start mechanism collects a minimal number (1 to 5) of samples through a targeted physical trial-cutting process for online fine-tuning, aligning the model with the current physical environment. Experiments on the PHM2010 dataset fully simulate varying cutting scenarios. The experimental results demonstrate that the proposed framework consistently outperforms traditional transfer learning, deep learning models, and existing meta-learning approaches, offering an effective solution for fast and accurate tool wear prediction under few-shot and varying cutting conditions.
Förderung
- Shenzhen Polytechnic University
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Shenzhen Strategic Emerging Industry Support Plans
- Shenzhen Sustainable Development Sci-Tech projec
Zitieren
APA
Zhou, J., Wang, L., & Wang, T. (2026). Tool Wear Prediction Under Varying Cutting Conditions: A Few-Shot Warm-Start Framework Based on Model-Agnostic Meta-Learning. Machines, 14(5), 471. https://doi.org/10.3390/machines14050471
Vancouver
Zhou J, Wang L, Wang T. Tool Wear Prediction Under Varying Cutting Conditions: A Few-Shot Warm-Start Framework Based on Model-Agnostic Meta-Learning. Machines. 2026;14(5):471. doi:10.3390/machines14050471
BibTeX
@article{Zhou2026,
author = {Zhou, Ju and Wang, Lin and Wang, Tao},
title = {Tool Wear Prediction Under Varying Cutting Conditions: A Few-Shot Warm-Start Framework Based on Model-Agnostic Meta-Learning},
journal = {Machines},
year = {2026},
volume = {14},
number = {5},
pages = {471},
doi = {10.3390/machines14050471}
}
EUVIMED