为解决燃煤电厂中选择性催化还原烟气脱硝系统（Selective Satalytic Reduction，SCR）入口NOx浓度测量延迟较大及难以准确测量的问题。本文基于某在役600MW超临界机组，首先提出基于互信息系数的动态延迟时间分析法，分析输入特征对NOx浓度的延迟影响；其次提出一种基于时序卷积网络与双向门控循环单元融合注意力机制（TCN-BiGRU-Attention）和残差修正的预测方法，用于燃煤电厂SCR入口NOx浓度预测。实验结果表明，提出的动态延迟分析法能够捕捉输入特征的最优时序映射关系，提升模型的预测效果。所提出的预测方法与传统LSTM、TCN-GRU等模型对比，在预测精度和鲁棒性方面具有显著优势。证明本文基于考虑输入特征对预测结果的延迟影响，提出的TCN-BiGRU-Attention与残差修正的预测方法，能够实现对SCR入口NOx浓度的准确预测，为脱硝技术提供有效指导。

In order to solve the problem of large delay and difficulty in accurate measurement of NOx concentration at the inlet of Selective Catalytic Reduction flue gas denitrification system (SCR) in coal-fired power plants, this paper proposes, firstly, a dynamic delay time analysis method based on mutual information coefficients to analyze the effect of input characteristics on NOx concentration. In this paper, based on an in-service 600MW supercritical unit, we firstly propose a dynamic delay time analysis method based on mutual information coefficients to analyze the delay effect of input features on NOx concentration; secondly, we propose a prediction method based on the fusion of time-sequence convolutional network and bi-directionally gated recurrent unit with the attention mechanism (TCN-BiGRU-Attention) and the residual correction for the prediction of inlet NOx concentration in SCR of a coal-fired power plant. Inlet NOx concentration prediction. The experimental results show that the proposed dynamic delay analysis method is able to capture the optimal temporal mapping relationship of the input features and improve the prediction effect of the model. The proposed prediction method has significant advantages in terms of prediction accuracy and robustness when compared with traditional models such as LSTM and TCN-GRU. It is demonstrated that the proposed prediction method of TCN-BiGRU-Attention with residual correction based on considering the delayed effect of input features on the prediction results can realize accurate prediction of NOx concentration at the SCR inlet and provide effective guidance for denitrification technology.

TK221

中国华能集团有限公司2022年度科技项目