Performance prediction and operating conditions optimization for aerobic fermentation heat recovery system based on machine learning

文献类型： 外文期刊

第一作者： He, Wei

作者： He, Wei;Cao, Yongna;Yu, Yanling;He, Wei;Cao, Yongna;Yu, Yanling;Qin, Jiang;Guo, Chao;Pei, Zhanjiang

作者机构：

关键词： Aerobic fermentation heat; Heat recovery; Machine learning; Performance prediction model; Operating conditions optimization

期刊名称：RENEWABLE ENERGY （ 影响因子：9.1； 五年影响因子：8.3 ）

ISSN： 0960-1481

年卷期： 2025 年 239 卷

页码：

收录情况： SCI

摘要： The low-grade, long-term production characteristics of aerobic fermentation heat (AFH) present challenges in heat recovery, leading to oversight. Clarifying the correlation between the operating conditions and performance of the aerobic fermentation heat recovery system (AFHRS) is crucial for achieving efficient AFH utilization. This study introduced machine learning tools to design and optimize an AFHRS that is suitable for two types of cold sources. A performance prediction model for the AFHRS was established with an average relative error of 3.66 %. The results demonstrate that, higher fermentation temperatures and flow rates exhibit advantageous effects on heat recovery performance. A moderate water flow is suitable for daily heating, providing a higher heat transfer rate (1573 W, 27.4 degrees C); a lower water flow is suitable for supplying hot water, providing a higher water temperature (898.7 W, 45.5 degrees C). High air flow shows unique potential in bio-drying, providing hot outdoor air (48.3 degrees C). In comparison to air (<= 18.8 %), water demonstrates satisfactory heat recovery efficiency (49.5 %- 82.1 %). However, maintaining optimum operating conditions for an extended duration is impractical. Intermittent heat recovery is a rational approach to align with the AFH production law. We deduced that cascaded heat exchangers can significantly increase AFH utilization efficiency.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序