Segmented Mechanism-Data Hybrid Modeling for Outlet SO<sub>2</sub> Mass Concentration Prediction in WFGD Systems Under Flexible Peak-shaving Conditions

LIU Chuang; CAO Wenguang; WANG Xiang; ZHANG Wenhui; LIU Guanqing

doi:10.19805/j.cnki.jcspe.2026.250341

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Segmented Mechanism-Data Hybrid Modeling for Outlet SO₂ Mass Concentration Prediction in WFGD Systems Under Flexible Peak-shaving Conditions

更新时间：2026-06-11

- Segmented Mechanism-Data Hybrid Modeling for Outlet SO₂ Mass Concentration Prediction in WFGD Systems Under Flexible Peak-shaving Conditions
- Chinese Journal of Power Engineering Vol. 46, Issue 6, Pages: 158-167(2026)
- 作者机构：
  
  1. 长沙理工大学能源与动力工程学院,湖南,长沙,410114
  2. 中国科学院过程工程研究所多相复杂系统国家重点实验室,北京,100190
  3. 清华大学能源与动力工程系,北京,100084
- 作者简介：
- 基金信息：
- DOI：10.19805/j.cnki.jcspe.2026.250341
  CLC： X701.3
- Published：2026
- 稿件说明：
移动端阅览
刘闯, 曹文广, 王翔, 张文慧, 柳冠青. 灵活调峰工况下WFGD系统出口SO2质量浓度的分段机理-数据混合建模[J]. 动力工程学报, 2026, 46(6): 158-167.
刘闯, 曹文广, 王翔, 张文慧, 柳冠青. 灵活调峰工况下WFGD系统出口SO2质量浓度的分段机理-数据混合建模[J]. 动力工程学报, 2026, 46(6): 158-167. DOI： 10.19805/j.cnki.jcspe.2026.250341.

DOI： 10.19805/j.cnki.jcspe.2026.250341

摘要

考虑火电机组灵活调峰运行过程中关键运行参数的动态变化

提出一种融合分段策略和机理-数据混合驱动的湿法烟气脱硫(WFGD)系统出口SO2质量浓度建模方法。通过分析各关键运行参数在调峰运行过程中的变化特征

将原始样本划分为深度调峰区间与基本调峰区间

并在各区间内分别构建机理模型、时序卷积网络-长短期记忆网络(TCN-LSTM)数据驱动模型及机理-数据混合模型。结果表明:分段建模策略在提高出口SO2质量浓度建模精度方面具有显著优势

且对不同建模方法均具有良好的适应性

其中机理-数据混合模型表现最优

其RMAPE和RMAE相比于原始全工况建模误差最大降低了33.4%和42.1%。

Abstract

Considering the dynamic change of key operating parameters during flexible peak-shaving operation process of coal-fired units

a modeling approach for outlet SO2 mass concentration in WFGD systems was proposed

which integrated a segmentation strategy with mechanism-data hybrid modeling. Based on the analysis of variation characteristics of key operating parameters during peak-shaving operation process

the dataset was divided into deep and basic peak-shaving intervals

within which mechanism models

temporal convolutional network-long short term memory (TCN-LSTM) models

and hybrid models were built. Results show that segmented modeling strategy has significant advantages in enhancing modeling accuracy of outlet SO2 mass concentration and has good adaptability to different modeling approaches. The mechanism-data hybrid model achieves the best performance

with RMAPE and RMAE reduced by up to 33.4% and 42.1%

respectively

compared with original full-condition modeling.

关键词

Keywords

references

孙月巧, 郑宏飞, 孔慧. 碳中和背景下煤电转型关键技术研究与展望[J]. 动力工程学报, 2022, 42(11): 1013-1023. SUN Yueqiao, ZHENG Hongfei, KONG Hui. Key technologies and prospects of coal power transformation under carbon neutrality background[J]. Journal of Chinese Society of Power Engineering, 2022, 42(11): 1013-1023.

LIU Quanbo, LI Xiaoli, WANG Kang. Real-time dynamic modelling of industrial WFGD process using an intelligence-based multi-model approach[J]. Measurement, 2025, 249: 117051.

范海东, 陈竹, 赵中阳, 等. 1000 MW燃煤机组湿法脱硫装置氧化系统运行优化[J]. 浙江大学学报(工学版), 2023, 57(4): 675-682. FAN Haidong, CHEN Zhu, ZHAO Zhongyang, et al. Operating optimization of oxidation subsystem of wet flue gas desulfurization system of 1000 MW coal-fired unit[J]. Journal of Zhejiang University (Engineering Science), 2023, 57(4): 675-682.

刘泉伯, 李晓理, 王康. 基于深度AttLSTM网络的脱硫过程建模研究[J]. 北京工业大学学报, 2024, 50(2): 140-151. LIU Quanbo, LI Xiaoli, WANG Kang. Desulfurization process modeling by using deep AttLSTM-based network[J]. Journal of Beijing University of Technology, 2024, 50(2): 140-151.

LI Ruilian, ZENG Deliang, LI Tingting, et al. Real-time prediction of SO2 emission concentration under wide range of variable loads by convolution-LSTM VE-transformer[J]. Energy, 2023, 269: 126781.

金秀章, 刘岳, 于静, 等. 基于变量选择和EMD-LSTM网络的出口SO2浓度预测[J]. 中国电机工程学报, 2021, 41(24): 8475-8483. JIN Xiuzhang, LIU Yue, YU Jing, et al. Prediction of outlet SO2concentration based on variable selection and EMD-LSTM network[J]. Proceedings of the CSEE, 2021, 41(24): 8475-8483.

XU Yueyang, WANG Yibo, LIU Chuang, et al. Adaptive robust optimal scheduling of combined heat and power microgrids based on photovoltaic mechanism/data fusion-driven power prediction[J]. Energies, 2025, 18(3): 732.

WANG Jinjiang, LI Yilin, GAO R X, et al. Hybrid physics-based and data-driven models for smart manufacturing: modelling, simulation, and explainability[J]. Journal of Manufacturing Systems, 2022, 63: 381-391.

ZHAO Zhongyang, FAN Haidong, LI Qinwu, et al. Hybrid modeling and real-time predictive scheduling of wet flue gas desulfurization for energy saving and life extension[J]. Energy & Fuels, 2023, 37(7): 5312-5322.

李瑞连, 曾德良, 张光明, 等. 基于机理和数据混合模型的SD-WFGD系统SO2浓度动态预测[J]. 动力工程学报, 2024, 44(7): 1118-1128, 1136. LI Ruilian, ZENG Deliang, ZHANG Guangming, et al. Dynamic prediction for SO2 concentration in SD-WFGD system based on mechanism-data combined model[J]. Journal of Chinese Society of Power Engineering, 2024, 44(7): 1118-1128, 1136.

张少强, 陈露, 刘子易, 等. 大型燃煤锅炉深度调峰关键问题探讨[J]. 南方能源建设, 2022, 9(3): 16-28. ZHANG Shaoqiang, CHEN Lu, LIU Ziyi, et al. Discussion on key problems of depth peak adjustment for large coal-fired boilers[J]. Southern Energy Construction, 2022, 9(3): 16-28.

李明扬, 蒋媛媛. 考虑煤耗率的火电机组灵活调峰对风电消纳的影响效果研究[J].热力发电,2020,49(2):45-51. LI Mingyang, JIANG Yuanyuan. Effect of flexible load regulation of thermal power units considering coal consumption rate on wind power utilization[J]. Thermal Power Generation,2020,49(2):45-51.

XU Changwei, NIE Wen, LIU Fei, et al. Improvement and optimization of coal dust concentration detection technology: based on the 3 criterion and the Kalman filtering composite algorithm[J]. Flow Measurement and Instrumentation, 2024, 97: 102598.

田亮, 王冠杰. 基于模型-数据联合的光伏-光热系统储能量预测[J]. 动力工程学报, 2024, 44(6): 911-918. TIAN Liang, WANG Guanjie. Energy storage prediction of photovoltaic-concentrating solar power system based on model and data[J]. Journal of Chinese Society of Power Engineering, 2024, 44(6): 911-918.

WANG Fang, MA Suxia, WANG He, et al. Prediction of NOx emission for coal-fired boilers based on deep belief network[J]. Control Engineering Practice,2018,80:26-35.

吴军英, 路欣, 刘宏, 等. 基于Spearman-GCN-GRU模型的超短期多区域电力负荷预测[J]. 中国电力, 2024, 57(6): 131-140. WU Junying, LU Xin, LIU Hong, et al. Ultra-short-term multi-region power load forecasting based on Spearman-GCN-GRU model[J]. Electric Power, 2024, 57(6): 131-140.

许大通. 超超临界燃煤发电机组深度调峰价值工程及运行策略研究[D]. 南京: 东南大学, 2021.

GAO Wei, LIU Ming, XIN Haozhe, et al. Control strategy optimization for wet flue gas desulfurization system during load cycling dynamic processes: energy saving and environmental impact[J]. Energy, 2024, 303: 131909.

ZHANG Yang, BOCQUET M, MALLET V, et al. Real-time air quality forecasting, part I: history, techniques, and current status[J]. Atmospheric Environment, 2012, 60: 632-655.

张军. 超低排放的湿法高效脱硫协同除尘的机理及模型研究[D]. 杭州: 浙江大学, 2018.

LEA C, FLYNN M D, VIDAL R, et al. Temporal convolutional networks for action segmentation and detection[C] //Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, Hawaii, USA: IEEE, 2017: 1003-1012.

YANG Tingting, MA Kangfeng, LÜ You, et al. Real-time dynamic prediction model of NOx emission of coal-fired boilers under variable load conditions[J]. Fuel, 2020, 274: 117811.

唐振浩, 柴向颖, 曹生现, 等. 考虑时延特征的燃煤锅炉NOx排放深度学习建模[J]. 中国电机工程学报, 2020, 40(20): 6633-6643. TANG Zhenhao, CHAI Xiangying, CAO Shengxian, et al. Deep learning modeling for the NOx emissions of coal-fired boiler considering time-delay characteristics[J]. Proceedings of the CSEE, 2020, 40(20): 6633-6643.

GONG Jianqiang, QU Zhiguo, ZHU Zhenle, et al. Ensemble models of TCN-LSTM-LightGBM based on ensemble learning methods for short-term electrical load forecasting[J]. Energy, 2025, 318: 134757.

周新宇, 姜志航, 白峻铭, 等. 基于IDBO-TCN-LSTM的短期光伏功率预测[J]. 计算机技术与发展, 2025, 35(7): 125-132. ZHOU Xinyu, JIANG Zhihang, BAI Junming, et al. Short-term PV power prediction based on IDBO-TCN-LSTM[J]. Computer Technology and Development, 2025, 35(7): 125-132.

王琦, 柴宇唤, 王鹏程, 等. 基于改进INFO-Bi-LSTM模型的SO2排放质量浓度预测[J]. 动力工程学报, 2024, 44(4): 641-649. WANG Qi, CHAI Yuhuan, WANG Pengcheng, et al. Prediction of SO2 emission mass concentration based on improved INFO-Bi-LSTM algorithm[J]. Journal of Chinese Society of Power Engineering, 2024, 44(4): 641-649.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Modeling and Performance Analysis of Molten Salt Energy Storage System in the Flexibility Retrofit of Coal-fired Power Plant

Thermodynamic Performance Analysis of Membrane Absorption CO2 Capture System for Coal-fired Power Plants

Experimental Study on Particle Concentration Attenuation Coefficient in the Riser of a Circulating Fluidized Bed

Distributionally Robust Optimal Dispatching of Integrated Energy System Considering Power-to-Ammonia and Flexible Coal-fired Generation

Research on Short-term Optimal Scheduling Strategy of Wind-PV- Thermal-BESS Considering Wind-PV Uncertainty

Related Author

No data

Related Institution

No data

AI问答

⁰

Segmented Mechanism-Data Hybrid Modeling for Outlet SO2 Mass Concentration Prediction in WFGD Systems Under Flexible Peak-shaving Conditions

DOI：10.19805/j.cnki.jcspe.2026.250341

摘要

Abstract

关键词

Keywords

references

Segmented Mechanism-Data Hybrid Modeling for Outlet SO₂ Mass Concentration Prediction in WFGD Systems Under Flexible Peak-shaving Conditions