机器学习辅助模型用于预测在各种原料和操作条件下油性污泥热解
Cheng Lu1, Dixuan Li1, Beidou Xi2
1Environmental Engineering Program, University of Northern British Columbia, Prince George, British Columbia V2N 4Z9, Canada.
Journal of hazardous materials
|February 21, 2025
概括
机器学习 (ML) 通过预测结果来优化油性污泥热解. 一个极端梯度增强 (XGB) 模型确定了泥灰,含量和温度等关键因素,以有效地回收资源和处理残留物.
科学领域:
- 化学工程是化学工程的重要组成部分.
- 环境科学 环境科学
- 数据科学数据科学数据科学
背景情况:
- 油性污泥热解提供了资源回收和安全的残留物处置.
- 热解的实验优化是昂贵和耗时的.
- 机器学习为流程优化提供了一个可行的替代方案.
研究的目的:
- 开发和验证用于预测和优化油性污泥热解的机器学习模型.
- 为了确定影响热解性能的关键因素.
- 为了减少与实验优化相关的时间和成本.
主要方法:
- 评估了六种机器学习模型,其中XGB (极端梯度提升) 由于其卓越的预测准确性而被选中.
- 一个多任务的XGB模型是使用油性污泥的最终/近似组成和热解运行条件作为输入来构建的.
- 对模型的性能进行了验证,平均R平方值为0.90.
主要成果:
- 该XGB模型准确地预测了油性污泥热解性能.
- 泥灰和含量以及热解温度被确定为最有影响力的因素.
- 最终成分 (42.5%),近距离特性 (35.8%) 和操作条件 (21.7%) 显著影响了热解结果.
结论:
- 机器学习,特别是XGB,为了解和优化油性污泥热解提供了有效的工具.
- 这种方法有助于有效的资源回收和废物管理.
- 开发的模型为工业应用提供了宝贵的见解,减少了实验负担.
相关概念视频
Microbial Bioremediation of Hydrocarbons
150
Bioremediation is an environmentally sustainable process that employs living organisms—primarily microorganisms—to degrade or neutralize pollutants from contaminated environments. In oil spills and hydrocarbon pollution, bioremediation involves the use of hydrocarbon-degrading bacteria to transform toxic compounds into less harmful substances. This approach leverages natural microbial metabolic processes and is considered both cost-effective and ecologically favorable compared to...
150
Biofuels
107
The microbial conversion of organic matter into biofuels holds potential as a renewable energy source. Among biofuel sources, microalgae are recognized as a highly efficient and adaptable feedstock for biodiesel production, owing to their rapid biomass accumulation, elevated lipid productivity, and capacity to proliferate in diverse aquatic systems, including freshwater, marine, and wastewater habitats. Unlike terrestrial crops, microalgae do not compete for land and can achieve significantly...
107


