期刊文献+

机器学习算法在脱硫系统智能运行及优化中的应用

Application of machine learning algorithms in intelligent operation of desulfurization systems
下载PDF
导出
摘要 随人工智能和大数据的飞速发展,以机器学习和工业大数据为核心的智能化系统已成为发电企业发展趋势。通过建立脱硫系统数据模型,结合机器学习算法自学习、自适应的特性,借助系统特征数据采集和上传功能,实现脱硫系统的自我诊断、提前预警及在线优化,有助于提升电厂智能化水平,达到超低排放、低碳运行的双重目标。在脱硫系统数据模型中,核心算法的选择将直接影响参数预测及后续优化的精准性。介绍了支持向量机、集成学习算法、神经网络法等主流机器学习算法的基本原理,综述了不同算法在脱硫系统智能运行中的应用,对比分析了不同算法在实现脱硫系统关键参数预测时的优缺点以及脱硫系统智能运行、优化方法。基于此确立了适用脱硫系统智能运行的更精准预测模型与优化技术路线,为实现湿法脱硫超低排放条件下的低碳节能运行提供参考。 With the rapid development of artificial intelligence and big data,intelligent systems with machine learning and industrial big data as the core have become the mainstream trend in the development of power generation enterprises.By establishing the data model of desulfurization system,combining the self-learning and self-adaptive characteristics of machine learning algorithm,and realizing self-diagnosis,early warning and online optimization of desulfurization system with the function of system characteristic data collection and uploading,it helps to improve the intelligence level of power plants and achieve the dual goals of ultra-low emission and low-carbon operation.In the FGD system data model,the selection of core algorithms will directly affect the accuracy of parameter prediction and subsequent optimization.The basic principles of mainstream machine learning algorithms such as support vector machines were introduced,integrated learning algorithms neural network methods and the application of different algorithms in the intelligent operation of FGD systems were reviewed,and the advantages and disadvantages of different algorithms in realizing the prediction of key parameters of FGD systems and the FGD system was compared and analyzed.The advantages and disadvantages of different algorithms in achieving the prediction of key parameters of FGD system and the method of FGD system intelligent operation and optimization were compared and analyzed.Based on this,a more accurate prediction model and optimization technology route for the intelligent operation of FGD system is established,which provides a reference for the realization of low carbon and energy-saving operation under the condition of ultra-low emission of wet FGD.
作者 孔若琪 崔琳 董勇 KONG Ruoqi;CUI Lin;DONG Yong(National Engineering Laboratory for Reducing Emissions from Coal Combustion,Shandong University,Jinan 250061,China)
出处 《洁净煤技术》 CAS CSCD 北大核心 2023年第S02期406-414,共9页 Clean Coal Technology
基金 国家重点研发计划资助项目(2017YFF0209803) 山东省重大科技创新工程资助项目(2020CXGC011402)
关键词 机器学习算法 预测模型 脱硫系统 智能运行 machine learning algorithms predictive models desulfurization systems intelligent operation
  • 相关文献

参考文献25

二级参考文献286

共引文献157

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部