基于聚类算法与贝叶斯优化的BP神经网络实时碳排放量预测模型

姜宇恒; 乔宗良; 李逗; 任少君; 司风琪

doi:10.19666/j.rlfd.202502015

English

您当前的位置：

首页 >

文章列表页 >

基于聚类算法与贝叶斯优化的BP神经网络实时碳排放量预测模型

热力发电 2025年54卷第11期页码：126-135

热能科学研究 | 更新时间：2026-08-11

- 基于聚类算法与贝叶斯优化的BP神经网络实时碳排放量预测模型
- Real-time prediction model for carbon emission using BP neural network based on clustering algorithm and Bayesian optimization
- 作者机构：
  
  1. 东南大学能源与环境学院
  2. 江苏方天电力有限公司
- 作者简介：
- 基金信息：
- DOI：10.19666/j.rlfd.202502015
  中图分类号： TP18;X773
- 纸质出版：2025
- 稿件说明：
移动端阅览
[1]姜宇恒,乔宗良,李逗,等.基于聚类算法与贝叶斯优化的BP神经网络实时碳排放量预测模型[J],2025,54(11):126-135.
[1]姜宇恒,乔宗良,李逗,等.基于聚类算法与贝叶斯优化的BP神经网络实时碳排放量预测模型[J],2025,54(11):126-135. DOI： 10.19666/j.rlfd.202502015.

DOI：10.19666/j.rlfd.202502015.

摘要

为了构建燃煤电厂碳排放量预测模型，针对燃煤机组普遍缺少入炉煤实时元素分析的问题，依据某百万千瓦机组2023年入炉煤质信息，以低位发热量、挥发分、硫分作为煤质划分依据，采用K-means++算法进行聚类分析，通过相关性分析筛选碳排放量预测模型的输入参数，基于贝叶斯优化的BP神经网络对聚类后各簇数据分别构建碳排放量预测模型，并对升降负荷等工况进行模型测试。结果显示，经煤质聚类后的模型在预测碳排放量时准确性显著提高，与未聚类的模型相比，平均均方根误差和平均相对误差最优情况降低约53.4%、49.2%，特别是在变负荷工况下，预测结果较准确。该方法不仅能有效预测燃煤电厂的碳排放量，还能在煤质复杂多变的情况下保持较高精度。

Abstract

To construct a prediction model for carbon emission from coal-fired power plants and address the problem of general lack of real-time elemental analysis for coal entering the furnace of coal-fired units

according to the in-furnace coal quality information of a million kilowatt unit in 2023

the low calorific value

volatile matter

and sulfur content were used as the basis for coal quality classification

K-means++ algorithm was used for clustering analysis

and correlation analysis was used to screen the input parameters of the carbon emission prediction model. The BP neural network suffered Bayesian optimization was used to construct carbon emission prediction models for each cluster data after clustering

and the models were tested for working conditions such as load increase and decrease. The results show that

the accuracy of the coal quality clustering model in predicting carbon emissions increases significantly. Compared with the non clustered model

the optimal cases of average root mean square error and average relative error reduce by about 53.4% and 49.2%

respectively. Especially under variable load conditions

the predicted results are more in line with the actual values. This indicates that the proposed method can not only effectively predict the carbon emissions of coal-fired power plants

but also maintain high accuracy in the case of complex and variable coal quality.

关键词

Keywords

references

WANG S, LIN B. Performance linkage in renewable energy supply chain:a comparative analysis with coal power and the entire industry[J]. Energy Policy, 2024,193:114274.

张小丽,崔学勤,王克,等.中国煤电锁定碳排放及其对减排目标的影响[J].中国人口·资源与环境, 2020,30(8):31-41.ZHANG Xiaoli, CUI Xueqin, WANG Ke, et al. China’s coal power locks in carbon emissions and its impact on emission reduction targets[J]. China's Population,Resources and Environment, 2020, 30(8):31-41

DU L, ZHAO H, TANG H, et al. Analysis of the synergistic effects of air pollutant emission reduction and carbon emissions at coal-fired power plants in China[J].Environmental Progress&Sustainable Energy, 2021,40(5):e13630.

陈怡,田川,曹颖,等.中国电力行业碳排放达峰及减排潜力分析[J].气候变化研究进展, 2020, 16(5):632-640.CHEN Yi, TIAN Chuan, CAO Ying, et al. Analysis of peak carbon emissions and emission reduction potential in China’s power industry[J]. Progress in Climate Change Research, 2020, 16(5):632-640

张丽.火电厂碳排放的计算方法及降低碳排放的有效调整方法[J].中外企业家, 2016(15):228.ZHANG Li. Calculation methods for carbon emissions in thermal power plants and effective adjustment methods for reducing carbon emissions[J]. Chinese and Foreign Entrepreneurs, 2016(15):228.

傅文明.火电厂锅炉混煤掺烧技术及节能运行措施研究[J].自动化应用, 2024, 65(18):25-27.FU Wenming. Research on coal mixing and combustion technology and energy saving operation measures for thermal power plant boilers[J]. Automation Application,2024, 65(18):25-27.

安吉振,陈衡,乔世超,等.基于K-means聚类算法的电站煤场来煤堆放优化研究[J].热力发电, 2023,52(4):135-143.AN Jizhen, CHEN Heng, QIAO Shichao, et al. Research on optimization of coal storage in power plant coal yard based on K-means clustering algorithm[J]. Thermal Power Generation, 2023, 52(4):135-143.

曹歌瀚,黄亚继,徐文韬,等.基于机器学习的燃煤锅炉分工况建模与燃烧优化[J].锅炉技术, 2023, 54(5):41-47.CAO Gehan, HUANG Yaji, XU Wentao, et al. Modeling and combustion optimization of coal-fired boilers based on machine learning for different operating conditions[J].Boiler Technology, 2023, 54(5):41-47

文雯,刘文哲,肖祥武,等.基于大数据和并行随机森林算法火电机组供电煤耗计算模型[J].热力发电,2018, 47(9):9-14.WEN Wen, LIU Wenzhe, XIAO Xiangwu, et al. A coal consumption calculation model for power supply of thermal power units based on big data and parallel random forest algorithm[J]. Thermal Power Generation, 2018,47(9):9-14.

郭魁星,李姚旺,何晓宜,等.面向用电碳计量的备用碳表系统优化配置方法[J].电力系统自动化, 2025,49(6):1-13.GUO Kuixing, LI Yaowang, HE Xiaoyi, et al.Optimization configuration method for backup carbon meter system for electricity carbon measurement[J].Power System Automation, 2025, 49(6):1-13.

董锐锋,陈浩军,曹振谦,等.燃煤锅炉碳排放核查方法及对比[J].工业炉, 2021, 43(2):41-44.DONG Ruifeng, CHEN Haojun, CAO Zhenqian, et al.Methods and comparison of carbon emission verification for coal fired boilers[J]. Industrial Furnace, 2021, 43(2):41-44.

PROKHORSKII G, RUDRA S, PREI?INGER M, et al. A data-driven regression model for predicting thermal plant performance under load fluctuations[J]. Carbon Neutrality, 2024, 3(1):1-15.

袁鹏,谭琛,李锋,等.基于迁移深度强化学习的火电机组实时碳排放预测方法[J].工业加热, 2024, 53(7):65-69.YUAN Peng, TAN Chen, LI Feng, et al. Real time carbon emission prediction method for thermal power units based on transfer deep reinforcement learning[J]. Industrial Heating, 2024, 53(7):65-69.

WANG X, YAN C, LIU W, et al. Research on carbon emissions prediction model of thermal power plant based on SSA-LSTM algorithm with boiler feed water influencing factors[J]. Sustainability, 2022, 14:15988.

CHEN J, ZHENG L, CHE W, et al. A method for measuring carbon emissions from power plants using a CNN-LSTM-Attention model with Bayesian optimization[J]. Case Studies in Thermal Engineering,2024, 63:105334.

LU C, LI W, GAO S. Driving determinants and prospective prediction simulations on carbon emissions peak for China’s heavy chemical industry[J]. Journal of Cleaner Production, 2020, 251:119642.

卢青.基于改进的K-means聚类算法的火电厂锅炉燃烧优化研究[D].秦皇岛:燕山大学, 2015:1.LU Qing. Research on combustion optimization of thermal power plant boilers based on improved k-means clustering algorithm[D]. Qinhuangdao:Yanshan University, 2015:1.

ARTHUR D, VASSILVITSKII S. K-Means++:the advantages ofcareful seeding[C]//Proceedings of the 8th Annual ACM-SIAM Symposium on Discrete Algorithms.Philadelphia, PA:Society for Industrial and Applied Mathematics, 2007:1027-1035.

孙林,刘梦含,徐久成.基于优化初始聚类中心和轮廓系数的K-means聚类算法[J].模糊系统与数学,2022, 36(1):47-65.SUN Lin, LIU Menghan, XU Jiucheng. K-means clustering algorithm based on optimizing initial clustering centers and contour coefficients[J]. Fuzzy Systems and Mathematics, 2022, 36(1):47-65.

何选森,何帆,于海澜. K-均值算法的初始化改进与聚类质量评估[J].西安工程大学学报, 2024, 38(6):114-123.HE Xuansen, HE Fan, YU Hailan. Initialization improvement and clustering quality evaluation of Kmeans algorithm[J]. Journal of Xi’an Engineering University, 2024, 38(6):114-123.

王仲,顾煜炯,韩旭东,等.基于历史数据聚类的火电机组工况划分[J].仪器仪表学报, 2019, 40(2):90-95.WANG Zhong, GU Yujiong, HAN Xudong, et al.Classification of thermal power unit operating conditions based on historical data clustering[J]. Journal of Instrumentation and Design, 2019, 40(2):90-95.

安吉振.火电机组燃料系统运行优化及故障预警方法研究[D].北京:华北电力大学, 2023:1.AN Jizhen. Research on optimization of fuel system operation and fault warning methods for thermal power units[D]. Beijing:North China Electric Power University(Beijing), 2023:1.

雷萌,李翠,王鑫,等.基于模糊聚类径向基神经网络的煤质分析模型[J].选煤技术, 2015(4):10-14.LEI Meng, LI Cui, WANG Xin, et al. Coal quality analysis model based on fuzzy clustering radial basis function neural network[J]. Coal Preparation Technology, 2015(4):10-14.

贺思云,高建瓴,陈岚.基于改进人工蜂群算法的Kmeans聚类算法[J].贵州大学学报(自然科学版), 2017,34(5):83-87.HE Siyun, GAO Jianling, CHEN Lan. K-means clustering algorithm based on improved artificial bee colony algorithm[J]. Journal of Guizhou University(Natural Science Edition), 2017, 34(5):83-87.

龚广京,周光,郑涛,等.基于线性回归与BP神经网络的火电厂燃煤碳排放计算研究[J].热能动力工程,2024, 39(3):73-81.GONG Guangjing, ZHOU Guang, ZHENG Tao, et al.Research on carbon emission calculation of coal fired power plants based on linear regression and bp neural network[J]. Thermal Power Engineering, 2024, 39(3):73-81.

WANG G, JIA R, LIU J, et al. A hybrid wind power forecasting approach based on Bayesian model averaging and ensemble learning[J]. Renewable Energy, 2020, 145:2426-2434.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

关联资源

基于融合全局残差Mamba的短期电价预测方法研究

热工仪表非线性校正方法的研究

沙角A电厂5号锅炉低NOx排放分析

减小入厂煤采样系统误差

基于L-M算法的火电厂实时数据神经网络预测模型研究