Scholarlink

<p>This paper presents a predictive modeling framework for managing volatile organic compound (VOC) emissions in VAT photopolymerization (VPP) additive manufacturing, enabled by an industrial Internet of Things (IoT) sensing system and a hybrid CNN-LSTM deep learning model. Using an Anycubic Mono X 6K 3D printer machine equipped with Internet of Things (IoT) sensors, we designed a hybrid CNN-LSTM model to predict VOC emissions from core printing parameters. The model scored a coefficient of determination (<i>R</i><sup>2</sup> value) of 0.931. The analysis also revealed exposure time as the most influential factor (correlation coefficient: 0.879). Light intensity affected VOC release moderately, while layer thickness had little impact. A mobile application based on Flutter was developed for real-time monitoring and a decision-support platform. Users can track VOC levels, resin temperature, room temperature, and humidity alongside model-based emission predictions, enabling proactive adjustments to maintain safe emissions levels. The system allows for proactive adjustments, so operators can maintain safe emissions levels and increase operational efficiency. By integrating data-driven decision-making into additive manufacturing processes, this research moves toward sustainable manufacturing by demonstrating the feasibility of VPP additive manufacturing for greener and more efficient production practices.</p> Graphical Abstract <p></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

Toward eco-friendly additive manufacturing: proactive VOC emission management in VAT photopolymerization using deep learning-based predictive analytics

  • 期刊论文

  • 发表日期：

    2026年4月20日

查看全文
  • Dhal A. Matoc,
  • Bhavesh K. Kanabar,
  • Amit Sata

摘要

This paper presents a predictive modeling framework for managing volatile organic compound (VOC) emissions in VAT photopolymerization (VPP) additive manufacturing, enabled by an industrial Internet of Things (IoT) sensing system and a hybrid CNN-LSTM deep learning model. Using an Anycubic Mono X 6K 3D printer machine equipped with Internet of Things (IoT) sensors, we designed a hybrid CNN-LSTM model to predict VOC emissions from core printing parameters. The model scored a coefficient of determination (R2 value) of 0.931. The analysis also revealed exposure time as the most influential factor (correlation coefficient: 0.879). Light intensity affected VOC release moderately, while layer thickness had little impact. A mobile application based on Flutter was developed for real-time monitoring and a decision-support platform. Users can track VOC levels, resin temperature, room temperature, and humidity alongside model-based emission predictions, enabling proactive adjustments to maintain safe emissions levels. The system allows for proactive adjustments, so operators can maintain safe emissions levels and increase operational efficiency. By integrating data-driven decision-making into additive manufacturing processes, this research moves toward sustainable manufacturing by demonstrating the feasibility of VPP additive manufacturing for greener and more efficient production practices.

Graphical Abstract