Scholarlink

In Industry 4.0, optimizing production performance amidst varying feedstock properties is a key challenge. This paper presents a novel data-driven modeling approach for a distillation unit (DU), integrating feedstock property and production feature extraction. The proposed method addresses the issue of extracting meaningful features from high-dimensional, imperfect industrial data, where product quality data is often unavailable. By leveraging the dynamic characteristics of the process, the model captures feedstock properties in a data-driven, knowledge-oriented way. The PM-FP-PF model, designed with a customized network structure, effectively predicts product quality even with incomplete data. Experimental results show its strong generalization ability across different feedstocks, offering a solid foundation for optimizing industrial operations and improving production efficiency and product quality.

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

Data-Driven Modeling Integrating Feedstock Time Series Features for Industrial Process Quality Prediction

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sihong Li,
  • Xiaohong Yin,
  • Wentao Liu,
  • Kaili Yin,
  • Yingrui Zhou

摘要

In Industry 4.0, optimizing production performance amidst varying feedstock properties is a key challenge. This paper presents a novel data-driven modeling approach for a distillation unit (DU), integrating feedstock property and production feature extraction. The proposed method addresses the issue of extracting meaningful features from high-dimensional, imperfect industrial data, where product quality data is often unavailable. By leveraging the dynamic characteristics of the process, the model captures feedstock properties in a data-driven, knowledge-oriented way. The PM-FP-PF model, designed with a customized network structure, effectively predicts product quality even with incomplete data. Experimental results show its strong generalization ability across different feedstocks, offering a solid foundation for optimizing industrial operations and improving production efficiency and product quality.