Cutting Feature Extraction Method for Ultra-High Molecular Weight Polyethylene Fiber-Reinforced Concrete Based on Feature Classification and Improved Hilbert–Huang Transform
Ultra-high molecular weight polyethylene (UHMWPE) fiber-reinforced concrete (UHMWPE-FRC) is a hard–soft multiphase hybrid composite with exceptional toughness and impact resistance compared to conventional concrete. However, its cutting characteristics and processing performance have not been suffic...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | Buildings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-5309/15/8/1272 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Ultra-high molecular weight polyethylene (UHMWPE) fiber-reinforced concrete (UHMWPE-FRC) is a hard–soft multiphase hybrid composite with exceptional toughness and impact resistance compared to conventional concrete. However, its cutting characteristics and processing performance have not been sufficiently investigated, potentially causing accelerated saw blade wear, higher energy consumption, and poor cutting quality, thus increasing project costs and duration. In order to intelligently evaluate the performance of diamond saw blades when cutting UHMWPE-FRC, a feature extraction method, based on feature classification and an improved Hilbert–Huang transform (HHT), is proposed, which consider Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) and wavelet threshold de-noising. By conducting the cutting experiments, the cutting force was analyzed by the improved HHT, in terms of noise reduction and time-frequency. Five types of characteristics were preliminarily screened, including depth of cut (<i>a<sub>p</sub></i>), cutting speed (<i>V<sub>c</sub></i>), feed rate (<i>V<sub>f</sub></i>), concrete strength, and the type of concrete. A feature correlation analysis method for UHMWPE-FRC cutting, based on feature classification, is proposed. The five features were classified into continuous variable features and unordered categorical variable features; correlation analyses were carried out by Spearman correlation coefficient testing and Kruskal–Wallis and Dunn’s testing, respectively. It was found that the <i>a<sub>p</sub></i> and concrete strength exhibited a strong positive correlation with cutting force, making them the primary influencing factors. Meanwhile, the influence of aggregates on cutting force can be identified in the low-frequency range, while the influence of fibers can be identified in the high-frequency range. The feature classification-based correlation analysis effectively distinguishes the influence of <i>V<sub>c</sub></i> on cutting force. |
|---|---|
| ISSN: | 2075-5309 |