A Boiler Reheater Wall Temperature Prediction Model Based on CNN-BiLSTM-SSA

XU Shiming; HE Zhiqian; PENG Xianyong; SHANG Zhongbao; FAN Jingwei; WANG Junlue; QU Shuyang; LIU Yang; ZHOU Huaichun

doi:10.19805/j.cnki.jcspe.2026.240658

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

A Boiler Reheater Wall Temperature Prediction Model Based on CNN-BiLSTM-SSA

更新时间：2026-10-10

- A Boiler Reheater Wall Temperature Prediction Model Based on CNN-BiLSTM-SSA
- Chinese Journal of Power Engineering Vol. 46, Issue 1, Pages: 121-130(2026)
- 作者机构：
  
  1. 中国矿业大学低碳能源与动力工程学院,江苏,徐州,221116
  2. 华能营口热电有限责任公司,辽宁,营口,115003
  3. 江苏省智慧能源技术及装备工程研究中心,江苏,徐州,221116
- 作者简介：
- 基金信息：
- DOI：10.19805/j.cnki.jcspe.2026.240658
  CLC：
- Online First：15 January 2026，
  
  Published：15 January 2026
- 稿件说明：
移动端阅览
徐世明,何至谦,彭献永,商忠宝,范景玮,王俊略,曲舒杨,刘洋,周怀春. 基于CNN-BiLSTM-SSA的锅炉再热器壁温预测模型动力工程学报, 2026, 46(1): 121-130 https://doi.

org/10.19805/j.cnki.jcspe.2026.240658
徐世明,何至谦,彭献永,商忠宝,范景玮,王俊略,曲舒杨,刘洋,周怀春. 基于CNN-BiLSTM-SSA的锅炉再热器壁温预测模型动力工程学报, 2026, 46(1): 121-130 https://doi. DOI： 10.19805/j.cnki.jcspe.2026.240658.

org/10.19805/j.cnki.jcspe.2026.240658 DOI：

摘要

针对锅炉高温再热器壁温动态特点

提出了一种基于稀疏自注意力(SSA)、卷积神经网络(CNN)及双向长短期记忆神经网络(BiLSTM)相融合的再热器壁温软测量模型。首先

采用核主成分分析(KPCA)算法对原始候选变量进行筛选降维

选择前26个主成分变量作为模型的最终输入。其次

考虑利用CNN捕捉局部相关性

BiLSTM学习数据的长期序列依赖性的优势

使用卷积神经网络-双向长短期记忆神经网络(CNN-BiLSTM)捕捉时序数据中的短期和长期依赖关系

引入稀疏自注意力SSA机制

通过为不同特征部分分配自适应权重

从而增强CNN-BiLSTM模型的特征提取与建模能力

最后利用在役1 000 MW超超临界锅炉的历史数据进行仿真实验。结果表明:CNN-BiLSTM-SSA模型在高温再热器壁温预测中的均方根误差(RMSE)、平均绝对误差(MAE)及平均绝对百分比误差(MAPE)分别为4.92℃、3.81℃和0.624 1%

相应的指标均优于CNN、LSTM、BiLSTM、CNN-LSTM和CNN-BiLSTM模型。

Abstract

In response to the dynamic characteristics of the boiler's high-temperature reheater wall temperature

a soft-sensor model for reheater wall temperature was proposed

which integrates a sparse self-attention mechanism (SSA)

a convolutional neural network (CNN)

and a bidirectional long short-term memory network (BiLSTM). First

the kernel principal component analysis (KPCA) algorithm was employed to screen and reduce the dimensionality of the original candidate variables

and the top twenty-six principal components were selected as the final inputs for the model. Secondly

leveraging the advantages of CNN in capturing local correlations and BiLSTM in learning long-term sequential dependencies

the CNN-BiLSTM framework was used to capture both short-term and long-term dependencies in the time-series data. The SSA mechanism was introduced to enhance the feature extraction and modeling capabilities of the CNN-BiLSTM model by assigning adaptive weights to different feature components. Finally

simulation experiments were conducted using historical data from an in-service 1 000 MW ultra-supercritical boiler. Results show that for high-temperature reheater wall temperature prediction

the proposed CNN-BiLSTM-SSA model achieves a root mean square error (RMSE) of 4.92 ℃

a mean absolute error (MAE) of 3.81 ℃

and a mean absolute percentage error (MAPE) of 0.624 1%. These corresponding metrics are all superior to those of the CNN

LSTM

BiLSTM

CNN-LSTM

and CNN-BiLSTM models.

关键词

Keywords

references

王孟浩, 王衡, 陈朝柱, 等. 高参数大容量电站锅炉过热器再热器的超温问题和炉内壁温在线监测[J]. 锅炉技术, 2009, 40(1): 5-11.WANG Menghao, WANG Heng, CHEN Chaozhu, et al. The heated tube overheating problem of superheater and reheater in high parameter large capacity utility boilers and tube wall temperature online monitor[J]. Boiler Technology, 2009, 40(1): 5-11.

王孟浩, 王衡, 郑民牛. 超(超)临界锅炉炉外壁温测点的测量误差[J]. 中国电力, 2009, 42(2): 45-48.WANGMenghao, WANG Heng, ZHENG Minniu. Measuring errors of external wall temperature for supercritical and ultra-supercritical boilers[J]. Electric Power, 2009, 42(2): 45-48.

陈朝松, 张树林, 刘平元, 等. 优化壁温计算模型及其在电站锅炉壁温在线监测中的应用[J]. 动力工程, 2009, 29(9): 818-822.CHEN Chaosong, ZHANG Shulin, LIU Pingyuan, et al. Application of optimized wall temperature calculation model in wall temperature on-line monitoring of utility boiler[J]. Journal of Power Engineering, 2009, 29(9): 818-822.

余岳溪, 廖宏楷, 吴燕玲, 等. 基于热工水力耦合求解的电站锅炉炉管壁温研究[J]. 动力工程学报, 2015, 35(1): 1-7.YU Yuexi, LIAO Hongkai, WU Yanling, et al. Study on tube wall temperature of power plant boilers based on coupled thermal hydraulic analysis[J]. Journal of Chinese Society of Power Engineering, 2015, 35(1): 1-7.

喻聪, 司风琪, 李敏, 等. 基于燃烧与水动力耦合模型的锅炉蒸汽管超温特性研究[J]. 热能动力工程, 2021, 36(8): 92-98.YU Cong, SI Fengqi, LI Min, et al. Study on overtemperature of boiler steam tube base on coupling model of combustion and hydrodynamic[J]. Journal of Engineering for Thermal Energy and Power, 2021, 36(8): 92-98.

金东昊, 刘欣, 张效源, 等. 燃煤锅炉屏式过热器壁温耦合计算方法[J]. 中国电机工程学报, 2022, 42(24): 8951-8960.JIN Donghao, LIU Xin, ZHANG Xiaoyuan, et al. A coupled model for predicting the tube temperature of platen superheater of coal-fired boiler[J]. Proceedings of the CSEE, 2022, 42(24): 8951-8960.

JIN Donghao, YAN Jingwen, LIU Xin, et al. Prediction of tube temperature distribution of boiler platen superheater by a coupled combustion and hydrodynamic model[J]. Energy, 2023, 279: 128116.

闫靖文, 刘欣, 王光礼, 等. 基于数据驱动的锅炉水冷壁壁温分布实时预测模型[J]. 煤炭学报, 2024, 49(10): 4117-4126.YAN Jingwen, LIU Xin, WANG Guangli, et al. Data-driven model for real-time prediction of waterwall tube temperature distributions of supercritical boiler[J]. Journal of China Coal Society, 2024, 49(10): 4117-4126.

MADEJSKI P, TALER D, TALER J. Thermal and flow calculations of platen superheater in large scale CFB boiler[J]. Energy, 2022, 258: 124841.

LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.

卢彬, 刘茜, 高林, 等. 基于NARX神经网络的锅炉壁温预测模型[J]. 热力发电, 2019, 48(3): 35-40.LU Bin, LIU Xi, GAO Lin, et al. Prediction model of boiler platen superheater tube wall temperature based on NARX neural network[J]. Thermal Power Generation, 2019, 48(3): 35-40.

CUI Zhipeng, XU Jing, LIU Wenhao, et al. Data-driven modeling-based digital twin of supercritical coal-fired boiler for metal temperature anomaly detection[J]. Energy, 2023, 278: 127959.

魏小兵, 崔智鹏, 徐婧, 等. 基于数据驱动的超临界锅炉受热面壁温预测方法[J]. 热力发电, 2023, 52(7): 106-112.WEI Xiaobing, CUI Zhipeng, XU Jing, et al. Data-driven method for predicting the wall temperature of heating surface of supercritical boilers[J]. Thermal Power Generation, 2023, 52(7): 106-112.

詹毅, 冯磊华, 杨锋, 等. 基于改进灰狼算法优化双向长短时记忆神经网络的水冷壁壁温预测[J]. 热力发电, 2024, 53(1): 188-196.ZHAN Yi, FENG Leihua, YANG Feng, et al. Prediction of water wall temperature based on improved grey wolf optimizer and bidirectionallong and short term memory network[J]. Thermal Power Generation, 2024, 53(1): 188-196.

曹熠云, 茅大钧, 陈思勤. 基于TCN-Attention-BiGRU的锅炉受热面壁温预测[J]. 计算机仿真, 2025, 42(7): 91-98.CAO Yiyun, MAO Dajun, CHEN Siqin. Boiler heating surface wall temperature prediction based on TCN-attention-BiGRU[J]. Computer Simulation, 2025, 42(7): 91-98.

NIU Dongxiao, YU Min, SUN Lijie, et al. Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism[J]. Applied Energy, 2022, 313: 118801.

吴铮, 张悦, 董泽. 基于多图卷积神经网络的主汽温系统故障诊断[J]. 动力工程学报, 2023, 43(2): 237-245.WU Zheng, ZHANG Yue, DONG Ze. Fault diagnosis of mainsteam temperature system based on multi-graph convolutional neural network[J]. Journal of Chinese Society of Power Engineering, 2023, 43(2): 237-245.

廖雪超, 陈才圣, 伍杰平. 基于CNN-LSTM及深度学习的风电场时空组合预测模型[J]. 信息与控制, 2022, 51(4): 498-512.LIAO Xuechao, CHEN Caisheng, WU Jieping. Combined spatiotemporal wind farm prediction model based on CNN-LSTM and deep learning[J]. Information and Control, 2022, 51(4): 498-512.

董坤, 冉鹏, 刘旭, 等. 基于权参数优化的并行深度学习光伏功率预测[J]. 动力工程学报, 2024, 44(1): 91-98.DONG Kun, RAN Peng, LIU Xu, et al. Photovoltaic power prediction by weight parameter optimization-based parallel deep learning[J]. Journal of Chinese Society of Power Engineering, 2024, 44(1): 91-98.

GREFF K, SRIVASTAVA R K, KOUTNK J, et al. LSTM: a search space odyssey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(10): 2222-2232.

SHERSTINSKY A. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network[J]. Physica D: Nonlinear Phenomena, 2020, 404: 132306.

CHEN Zhenghua, WU Min, ZHAO Rui, et al. Machine remaining useful life prediction via an attention-based deep learning approach[J]. IEEE Transactions on Industrial Electronics, 2021, 68(3): 2521-2531.

MA Xiaolei, ZHANG Jiyu, DU Bowen, et al. Parallel architecture of convolutional Bi-directional LSTM neural networks for network-wide metro ridership prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(6): 2278-2288.

VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, USA: Curran Associates Inc, 2017: 6000-6010.

罗毅, 段明达. 基于IWOA-Transformer的磨煤机故障预警[J]. 动力工程学报, 2024, 44(6): 939-946.LUO Yi, DUAN Mingda. Fault warning of coal mill based on IWOA-Transformer[J]. Journal of Chinese Society of Power Engineering, 2024, 44(6): 939-946.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Short-term Wind Power Output Prediction Based on FeatureSimilarity Day and CNN-BiLSTM-Attention Model

Prediction of NOx Concentration at the SCR Inlet Based on Improved LSTM with Attention Mechanism

Day-ahead Irradiance Forecasting Framework Based on CEEMDAN-KDBSCAN-Informer

Mooring-line Tension Prediction Based on Re-decomposition and Deep Learning

Research on Optimal Scheduling Method for Source-Grid- Load-Storage Systems Considering Electro-Thermal Load Demands in Industrial Parks

Related Author

LUAN Fuming

ZHANG Heng

CHEN Haiping

LIU Xubin

GUO Peng

LIN Feng

ZHAO Xudong

LI Jun

Related Institution

School of Energy Power and Mechanical Engineering, North China Electric Power University

North China Electric Power Test and Research Institute, China DatangCorporation Science and Technology General Research Institute Co., Ltd.

School of Control and Computer Engineering, North China Electric Power University

Xi'an Thermal Power Research Institute Co., LTD

National Key Laboratory of High-Efficiency Flexible Coal Power Generation and Carbon Capture Utilization and Storage

AI问答

⁰