Scholarlink

<p>Hybrid additive manufacturing (HAM) is characterized by process variability and low reproducibility, necessitating reliable real-time monitoring of tool condition during machining. In this study, tool condition monitoring was performed using weak acoustic emission (AE) signals acquired from a sensor remotely mounted on the spindle housing. Experiments were conducted using 18Ni-300 maraging steel fabricated via laser powder bed fusion. To detect tool condition changes under low signal-to-noise ratio (SNR) conditions caused by signal attenuation and rotational noise, a normalized RMS ratio index—defined as the ratio of hit RMS to pre-trigger RMS—was introduced, together with a threshold-based RMS ratio Count method. Based on experimental data collected from manufacturing cycles, the RMS ratio Count exhibited a pronounced increase at Cycle 9, which coincided with the occurrence of tool breakage. In particular, the number of events exceeding a threshold of 4 increased sharply at the failure cycle. For example, the RMS ratio Count (&gt; 4) increased from 13 to 20 at the breakage cycle. Furthermore, surface roughness measurements revealed a significant degradation after tool breakage, with the Sa value increasing from 1.37&#xa0;μm to 2.23&#xa0;μm (approximately 62.4%). These results demonstrate that the proposed RMS ratio Count is highly sensitive to tool fracture events, even under weak signal conditions. This study presents a normalization-based, high-sensitivity AE signal analysis approach for real-time tool breakage detection during the machining stage of HAM and demonstrates its applicability in low-SNR environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

RMS Ratio-based acoustic emission analysis for in-situ tool breakage detection in hybrid additive manufacturing

  • 期刊论文

  • 发表日期：

    2026年5月6日

查看全文
  • Hyokyu Kim,
  • Ohyung Kwon,
  • Dongwoon Han,
  • Dongjin Choi,
  • Seongtak Kim

摘要

Hybrid additive manufacturing (HAM) is characterized by process variability and low reproducibility, necessitating reliable real-time monitoring of tool condition during machining. In this study, tool condition monitoring was performed using weak acoustic emission (AE) signals acquired from a sensor remotely mounted on the spindle housing. Experiments were conducted using 18Ni-300 maraging steel fabricated via laser powder bed fusion. To detect tool condition changes under low signal-to-noise ratio (SNR) conditions caused by signal attenuation and rotational noise, a normalized RMS ratio index—defined as the ratio of hit RMS to pre-trigger RMS—was introduced, together with a threshold-based RMS ratio Count method. Based on experimental data collected from manufacturing cycles, the RMS ratio Count exhibited a pronounced increase at Cycle 9, which coincided with the occurrence of tool breakage. In particular, the number of events exceeding a threshold of 4 increased sharply at the failure cycle. For example, the RMS ratio Count (> 4) increased from 13 to 20 at the breakage cycle. Furthermore, surface roughness measurements revealed a significant degradation after tool breakage, with the Sa value increasing from 1.37 μm to 2.23 μm (approximately 62.4%). These results demonstrate that the proposed RMS ratio Count is highly sensitive to tool fracture events, even under weak signal conditions. This study presents a normalization-based, high-sensitivity AE signal analysis approach for real-time tool breakage detection during the machining stage of HAM and demonstrates its applicability in low-SNR environments.