School of Energy Science and Engineering, Central South University, Changsha 410083, China.
School of Minerals Processing and Bioengineering, Central South University, Changsha 410083, China.
Bioresour Technol. 2022 Oct;362:127791. doi: 10.1016/j.biortech.2022.127791. Epub 2022 Aug 17.
Hydrothermal liquefaction (HTL) of high-moisture biomass or biowaste to produce bio-oil is a promising technology. However, nitrogen-heterocycles (NH) presence in bio-oil is a bottleneck to the upgrading and utilization of bio-oil. The present study applied the machine learning (ML) method (random forest) to predict and help control the bio-oil NH, bio-oil yield, and N content of bio-oil (N_oil). The results indicated that the predictive performance of the yield and N_oil were better than previous studies, achieving test R of 0.92 and 0.95, respectively. Acceptable predictive performance (test R of 0.82 and RMSE of 7.60) for the prediction of NH was also achieved. The feature importance analysis, partial dependence, and Shapely value were used to interpret the prediction models and study the NH formation mechanisms and behavior. Then, forward optimization of NH was implemented based on optimal predictive models, indicating the high potential of ML-aided bio-oil production and engineering.
水热液化（HTL）高水分生物质或生物废物生产生物油是一项很有前途的技术。然而，生物油中氮杂环（NH）的存在是生物油升级和利用的瓶颈。本研究应用机器学习（ML）方法（随机森林）来预测和帮助控制生物油中的 NH、生物油产率和生物油中的 N 含量（N_oil）。结果表明，产率和 N_oil 的预测性能优于以往的研究，分别达到了测试 R 为 0.92 和 0.95。对 NH 的预测也取得了可接受的预测性能（测试 R 为 0.82 和 RMSE 为 7.60）。特征重要性分析、偏依赖和 Shapely 值用于解释预测模型，并研究 NH 的形成机制和行为。然后，基于最优预测模型进行 NH 的正向优化，表明 ML 辅助生物油生产和工程具有很高的潜力。
