摘要
随着人工智能技术的快速发展,利用先进算法解决传统的环保问题已成为电力行业发展的新趋势。综述了经典的机器学习和深度学习神经网络机理,详细阐述了神经网络算法在脱硫系统SO_(2)排放预测、pH预测和综合能效评价中的应用和发展前景。结果表明,LSTM深度神经网络在处理非线性回归、时序性、大惯性、延迟性等问题中具有突出优势,可应用于脱硫系统优化改造,并基于LSTM深度神经网络算法和环保设备的变频改造提出了燃煤电厂石灰石-石膏湿法精准脱硫方案。研究成果对实现脱硫系统的节能降耗,建设"智慧环保"和"数字环保"的燃煤电厂具有重要意义。
With the rapid development of artificial intelligence technology,the use of advanced algorithms to solve traditional environmental problems has become a new trend in electric power industry.The classical mechanisms of machine learning and deep learning neural network were summarized.The application and prospect of neural network algorithm in SO_(2)emission prediction,pH prediction and comprehensive energy efficiency evaluation of FGD system were elaborated in detail.Research results show that LSTM deep neural network has outstanding advantages in dealing with non-linear regression problems,as well as temporality,large inertia and lag objects like FGD system.Finally,based on LSTM deep neural network algorithm and frequency conversion transformation of equipment,limestone gypsum wet intelligent FGD scheme for coal-fired power plant was proposed.The research has great significance to realize energy saving and consumption reduction of FGD system,which are conducive to build intelligent and digital power plants.
作者
柴晋
乔加飞
孙灏
梁占伟
张千
CHAI Jin;QIAO Jiafei;SUN Hao;LIANG Zhanwei;ZHANG Qian(CHN Energy New Energy Technology Research Institute Co.,Ltd.,Beijing 102206,China)
出处
《洁净煤技术》
CAS
北大核心
2021年第S02期27-32,共6页
Clean Coal Technology
