Scholarlink

<p>Photoelectric fire smoke detectors are sensitive to false alarms caused by nuisance aerosols, which causes massive losses. To address this, techniques such as multiple optical channels and wavelengths have been developed to capture more particle scattering information. This paper presents a numerical simulation-based approach using broadband light (400–800&#xa0;nm) to capture multi-dimensional scattering information for particle discrimination. Using Mie scattering theory, we generated scattering spectrum for five types of fire smoke and five types of nuisance aerosols across various angles, which were then fed into five machine learning models for classification. By introducing random measurement noise, we tested model robustness. These results indicate that several forward scattering angles (45°–55°, 65°-75°) combined with nonlinear machine learning models like Random Forest and XGBoost achieved 100% precision and recall in discriminating fire smoke from nuisance aerosols. Additionally, the method accurately classified various fire smoke particle types with nearly 100% accuracy. This study highlights the potential of broadband visible light sources in fire detection, offering a robust solution to reduce false alarms and improve detection accuracy.</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

Simulations to Discriminate Between Different Fire Smokes and Nuisance Aerosols Through Broadband Light Scattering

  • 期刊论文

  • 发表日期：

    2025年7月25日

查看全文
  • Yuxin Kang,
  • Kaiyuan Li,
  • Weisi Liu,
  • Xiang Zhang,
  • Lida Huang,
  • Tao Chen,
  • Hongyong Yuan

摘要

Photoelectric fire smoke detectors are sensitive to false alarms caused by nuisance aerosols, which causes massive losses. To address this, techniques such as multiple optical channels and wavelengths have been developed to capture more particle scattering information. This paper presents a numerical simulation-based approach using broadband light (400–800 nm) to capture multi-dimensional scattering information for particle discrimination. Using Mie scattering theory, we generated scattering spectrum for five types of fire smoke and five types of nuisance aerosols across various angles, which were then fed into five machine learning models for classification. By introducing random measurement noise, we tested model robustness. These results indicate that several forward scattering angles (45°–55°, 65°-75°) combined with nonlinear machine learning models like Random Forest and XGBoost achieved 100% precision and recall in discriminating fire smoke from nuisance aerosols. Additionally, the method accurately classified various fire smoke particle types with nearly 100% accuracy. This study highlights the potential of broadband visible light sources in fire detection, offering a robust solution to reduce false alarms and improve detection accuracy.