基于深度学习的锅炉热效率动态预测模型研究

王志; 尹泳博; 徐世明; 彭献永; 周怀春

doi:10.19805/j.cnki.jcspe.2025.240549

欢迎来到能源电力期刊网！

English Version 读者登录期刊登录

您当前的位置：

首页 >

文章列表页 >

基于深度学习的锅炉热效率动态预测模型研究

数字化与智能化 | 更新时间：2025-11-28

- 基于深度学习的锅炉热效率动态预测模型研究
- Study on Dynamic Prediction Model of Boiler Thermal Efficiency Based on Deep Learning
- 动力工程学报 2025年45卷第11期页码：1892-1904
- 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
- 作者简介：
- 基金信息：
  
  江苏省卓越博士后计划资助项目(2024ZB049);国家重点研发计划资助项目(2023YFB4102800);国家自然科学基金重大科研仪器研制资助项目(51827808)
- DOI：10.19805/j.cnki.jcspe.2025.240549
  中图分类号：
- 网络出版：2025-11-17，
  
  纸质出版：2025-11-17
- 稿件说明：
移动端阅览
王志,尹泳博,徐世明,彭献永,周怀春. 基于深度学习的锅炉热效率动态预测模型研究动力工程学报, 2025, 45(11): 1892-1904 https://doi.

org/10.19805/j.cnki.jcspe.2025.240549
王志,尹泳博,徐世明,彭献永,周怀春. 基于深度学习的锅炉热效率动态预测模型研究动力工程学报, 2025, 45(11): 1892-1904 https://doi. DOI： 10.19805/j.cnki.jcspe.2025.240549.

org/10.19805/j.cnki.jcspe.2025.240549 DOI：

摘要

热效率是锅炉设备性能和机组运行经济性评价的重要指标。针对热效率与辅助变量之间的高维非线性

利用随机森林算法进行有监督降维

针对锅炉典型时序特性

根据适用于变负荷工况的卷积神经网络(convolutional neural network

CNN)构建热效率动态模型。为保证模型轻量化

在三层卷积层的常规CNN中引入通道均衡块以解决因特征值被抑制而存在的通道崩塌问题

构建了基于通道均衡卷积神经网络(channel equalization CNN

CE-CNN)的热效率预测模型

并在600 MW真实锅炉历史数据上开展了仿真实验。结果表明:CE-CNN的均方根误差为(0.1010.008)%

与6-layer CNN相比

节省了29.97%的训练时间

验证了模型的有效性。

Abstract

Thermal efficiency is an important indicator for the evaluation of boiler equipment performance and unit operation economy. For the high-dimensional nonlinearity between thermal efficiency and auxiliary variables

Random forest algorithm was used to carry out supervised dimensionality reduction

and the convolutional neural network (CNN) for variable load conditions was proposed to construct a dynamic model of thermal efficiency in view of the typical time-series characteristics of boiler. In order to ensure that the model was lightweight

a channel equalization block was introduced into the conventional CNN with three convolutional layers to solve the problem of channel collapse due to the suppression of feature maps

and a thermal efficiency prediction model based on channel equalization CNN (CE-CNN) was constructed. Simulation experiments were conducted using historical data from a 600 MW real boiler. Results show that the CE-CNN achieves a root mean square error of (0.1010.008)%

reducing training time by 29.97% compared to a 6-layer CNN

which validates the effectiveness of the model.

关键词

Keywords

references

ESLICK J C, ZAMARRIPA M A, MA Jinjiang, et al. Predictive modeling of a subcritical pulverized-coal power plant for optimization: parameter estimation, validation, and application[J]. Applied Energy, 2022, 319: 119226.

员盼锋, 徐舒涵, 丹慧杰, 等. 基于火电厂源侧的综合能源系统集成及优化配置研究[J]. 动力工程学报, 2024, 44(4): 650-657. YUAN Panfeng, XU Shuhan, DAN Huijie, et al. Integration and optimal allocation of integrated energy system based on source side of thermal power plant[J]. Journal of Chinese Society of Power Engineering, 2024, 44(4): 650-657.

ZHAO Huirong, SHEN Jiong, LI Yiguo, et al. Coal-fired utility boiler modelling for advanced economical low-NOx combustion controller design[J]. Control Engineering Practice, 2017, 58: 127-141.

喻聪. 电站锅炉燃烧优化及低NOx排放控制若干问题研究[D]. 南京: 东南大学, 2019.

DAL SECCO S, JUAN O, LOUIS-LOUISY M, et al. Using a genetic algorithm and CFD to identify low NOx configurations in an industrial boiler[J]. Fuel, 2015, 158: 672-683.

ZHOU Hao, CEN Kefa, MAO Jianbo. Combining neural network and genetic algorithms to optimize low NOx pulverized coal combustion[J]. Fuel, 2001, 80(15): 2163-2169.

SHI Yan, ZHONG Wenqi, CHEN Xi, et al. Combustion optimization of ultra supercritical boiler based on artificial intelligence[J]. Energy, 2019, 170: 804-817.

李杨, 蓝茂蔚, 赵国钦, 等. 基于PCA-PSO-LSSVM的电站锅炉效率预测模型研究[J]. 热力发电, 2021, 50(12): 43-50. LI Yang, LAN Maowei, ZHAO Guoqin, et al. Study on prediction model of utility boiler efficiency based on PCA-PSO-LSSVM[J]. Thermal Power Generation, 2021, 50(12): 43-50.

ZHAO Yufan, LIU Han, GUO Runyuan, et al. Air preheater rotor deformation soft sensor based on wavelet analysis and SVR[C]//Proceedings of the 2020 Chinese Automation Congress (CAC). Shanghai: IEEE, 2020: 4490-4495.

何宁, 谢天, 尹俊杰, 等. 燃煤机组变负荷瞬态过程的实时能耗分析[J]. 动力工程学报, 2023, 43(2): 158-164. HE Ning, XIE Tian, YIN Junjie, et al. Real-time energy consumption analysis of coal-fired power units during load change transient processes[J]. Journal of Chinese Society of Power Engineering, 2023, 43(2): 158-164.

文乐, 付龙龙, 杨新民. 超临界直流锅炉实际蓄热系数计算分析[J]. 热能动力工程, 2018, 33(12): 105-110. WEN Le, FU Longlong, YANG Xinmin. Calculation and analysis of actual heat storage coefficient for supercritical once-through boilers[J]. Journal of Engineering for Thermal Energy and Power, 2018, 33(12): 105-110.

ZHANG Y G. Combustion control of utility boiler based on GRU neural network[J]. International Core Journal of Engineering, 2022, 8(1): 665-667.

CHEN Wanghu, ZHAI Chenhan, WANG Xin, et al. GCN-and GRU-based intelligent model for temperature prediction of local heating surfaces[J]. IEEE Transactions on Industrial Informatics, 2023, 19(4): 5517-5529.

TUTTLE J F, BLACKBURN L D, ANDERSSON K, et al. A systematic comparison of machine learning methods for modeling of dynamic processes applied to combustion emission rate modeling[J]. Applied Energy, 2021, 292: 116886.

XIE Peiran, GAO Mingming, ZHANG Hongfu, et al. Dynamic modeling for NOx emission sequence prediction of SCR system outlet based on sequence to sequence long short-term memory network[J]. Energy, 2020, 190: 116482.

WANG Zhi, PENG Xianyong, ZHOU Huaichun, et al. A dynamic modeling method using channel-selection convolutional neural network: a case study of NOx emission[J]. Energy, 2024, 290: 130270.

LI Nan, HU Yong. The deep convolutional neural network for NOx emission prediction of a coal-fired boiler[J]. IEEE Access, 2020, 8: 85912-85922.

YANG Guotian, WANG Yingnan, LI Xinli. Prediction of the NOx emissions from thermal power plant using long-short term memory neural network[J]. Energy, 2020, 192: 116597.

TANG Zhehao, WANG Shikui, CHAI Xiangying, et al. Auto-encoder-extreme learning machine model for boiler NOx emission concentration prediction[J]. Energy, 2022, 256: 124552.

BREIMAN L. Bagging predictors[J]. Machine Learning, 1996, 24(2): 123-140.

VERIKAS A, GELZINIS A, BACAUSKIENE M. Mining data with random forests: a survey and results of new tests[J]. Pattern Recognition, 2011, 44(2): 330-349.

MEHTA D, KIM K I, THEOBALT C. On implicit filter level sparsity in convolutional neural networks[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 2019: 520-528.

IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille: JMLR. org, 2015: 448-456.

HUANG Lei, YANG Dawei, LANG Bo, et al. Decorrelated batch normalization[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 791-800.

KIM J H. Further improvement of Jensen inequality and application to stability of time-delayed systems[J]. Automatica, 2016, 64: 121-125.

YANG Jianwei, REN Zhile, GAN Chuang, et al. Cross-channel communication networks[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems. Vancouver: Curran Associates Inc., 2019: 1297-1306.

LEHMANN R. 3-rule for outlier detection from the viewpoint of geodetic adjustment[J]. Journal of Surveying Engineering, 2013, 139(4): 157-165.

JEON Y, KIM J. Active convolution: learning the shape of convolution for image classification[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1846-1854.

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.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向二次风精确描述的燃煤锅炉多维混合建模方法及变配风数值模拟研究