基于物理单调最小二乘支持向量机的锅炉燃烧NO<sub><i>x</i></sub>生成浓度预测方法

胡溆昂; 郭文康; 李益国

doi:10.19805/j.cnki.jcspe.2026.250044

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于物理单调最小二乘支持向量机的锅炉燃烧NO_x生成浓度预测方法

数字化与智能化 | 更新时间：2026-10-10

- 基于物理单调最小二乘支持向量机的锅炉燃烧NO_x生成浓度预测方法
- Prediction Method of Boiler Combustion NO_x Generation Concentration Based on Physics-constrained Monotonic Least Squares Support Vector Machine
- 动力工程学报 2026年46卷第5期页码：137-145
- 作者机构：
- 作者简介：
  
  胡溆昂(2000—),男,江苏徐州人,硕士研究生,研究方向为火电机组运行优化。李益国(通信作者),男,教授,博士,E-mail:lyg@seu.edu.cn。
- 基金信息：
  
  国家自然科学基金资助项目(51476027)
- DOI：10.19805/j.cnki.jcspe.2026.250044
  中图分类号： TK39;TM621
- 纸质出版：2026
- 稿件说明：
移动端阅览
胡溆昂, 郭文康, 李益国. 基于物理单调最小二乘支持向量机的锅炉燃烧NOx生成浓度预测方法[J]. 动力工程学报, 2026, 46(5): 137-145.
胡溆昂, 郭文康, 李益国. 基于物理单调最小二乘支持向量机的锅炉燃烧NOx生成浓度预测方法[J]. 动力工程学报, 2026, 46(5): 137-145. DOI： 10.19805/j.cnki.jcspe.2026.250044.

DOI： 10.19805/j.cnki.jcspe.2026.250044

摘要

火电机组是我国电力供应的基础

在电力生产中仍占主导地位。准确且符合物理约束的 NOx 浓度预测

对燃煤锅炉的燃烧优化、最佳运行工况保障及节能减排至关重要。机器学习可依托历史运行数据

为复杂的锅炉燃烧系统构建拟合精度高、计算速度快的 NOx生成浓度预测模型

但黑箱模型对训练数据依赖性强、可解释性不足

甚至可能与物理先验知识相悖。为此

提出一种基于物理单调性的自适应最小二乘支持向量机(MALSSVM)方法用于预测燃煤锅炉NOx生成浓度

将氧量、给煤量、燃尽风门开度与NOx生成浓度的物理先验单调性关系融入自适应最小二乘支持向量机(ALSSVM)中

以确保模型遵循物理约束

增强其在锅炉不同工况下的预测精度

并具备在对象特性变化时自动调整模型的能力。选取某660 MW燃煤锅炉开展案例研究。分析结果表明

该预测模型在测试集上的平均绝对预测误差(MAE)为20.84 mg/m3

性能提升约50%

可在未知工况下保持相关参数间的单调性关系

契合实际物理约束;在可解释性和泛化能力方面显著优于传统 ALSSVM 方法

能为火电机组 NOx 浓度精准预测提供更可靠的技术支撑。

Abstract

Thermal power units are the foundation of China's power supply and continue to play a dominant role in power production. Accurate prediction of NOx concentration that adheres to physical constraints is crucial for combustion optimization

ensuring optimal operating conditions

and achieving energy conservation and emission reduction in coal-fired boilers. Machine learning can leverage historical operating data to build NOx generation concentration prediction model with high fitting accuracy and fast calculation speed for complex boiler combustion systems. However

black-box models rely heavily on training data

lack interpretability

and may even contradict prior physical knowledge. To address this

a physics-constrained monotonic adaptive least squares support vector machine (MALSSVM) method was proposed for predicting NOx generation concentration in coal-fired boilers. The proposed method integrated the prior physical monotonic relationships between oxygen content

coal feed rate

overfire air damper opening

and NOx generation concentration into the adaptive least squares support vector machine (ALSSVM)

so as to ensure the model to follow physical constraints

enhance prediction accuracy under different boiler operating conditions

and possess the capability to automatically adjust the model when object characteristics changes. A case study was conducted on a 660 MW coal-fired boiler. The analysis results indicate that the mean absolute error (MAE) of the prediction model on the test set is 20.84 mg/m3

representing a performance improvement of approximately 50%. The model can maintain monotonic relationships among relevant parameters under unknown operating conditions

aligning with actual physical constraints. It is significantly superior to the traditional ALSSVM method in terms of interpretability and generalization capability

providing more reliable technical support for the accurate prediction of NOx concentration in thermal power units.

关键词

Keywords

references

MENG Xia, LIU Cong, CHEN Renjie, et al. Short term associations of ambient nitrogen dioxide with daily total, cardiovascular, and respiratory mortality: multilocation analysis in 398 cities[J]. BMJ, 2021, 372: n534.

国家标准. 火电厂大气污染物排放标准: GB 13223-2011[S]. 北京: 中国环境科学出版社, 2012.

严祯荣, 钟布依, 胡超, 等. 煤粉富氧燃烧CO2富集特性与NOx生成量模拟预测研究[J]. 动力工程学报, 2024, 44(4): 632-640.YAN Zhenrong, ZHONG Buyi, HU Chao, et al. Research on prediction of CO2 enrichment characteristics and NOx generation during oxygen-enriched combustion of pulverized coal[J]. Journal of Chinese Society of Power Engineering, 2024, 44(4): 632-640.

VIZGAVLJUST N V, STARCHENKO A V, GIL A V, et al. The processes of formation of nitrogen oxides in the boiler furnace BKZ 320-140[J]. EPJ Web of Conferences, 2015, 82: 01040.

张涛, 戴良旭, 王长安, 等. 燃煤耦合农林废弃物在流化床内燃烧的氮氧化物生成特性模拟研究[J]. 动力工程学报, 2024, 44(5): 692-702.ZHANG Tao, DAI Liangxu, WANG Chang' an, et al. Simulation study on nitrogen oxide formation characteristics in co-combustion of coal and agricultural waste in fluidized bed boiler[J]. Journal of Chinese Society of Power Engineering, 2024, 44(5): 692-702.

VAN DER LANS R P,GLARBORG P,DAM-JOHANSEN K.Influence of process parameters on nitrogen oxide formation in pulverized coal burners[J].Progress in Energy and Combustion Science, 1997, 23(4): 349-377.

唐振浩, 朱得宇, 李扬. 基于数据驱动的燃煤锅炉NOx排放浓度动态修正预测模型[J]. 中国电机工程学报, 2022, 42(14): 5182-5193.TANG Zhenhao, ZHU Deyu, LI Yang. Data driven based dynamic correction prediction model for NOx emission of coal fired boiler[J]. Proceedings of the CSEE, 2022, 42(14): 5182-5193.

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

方海泉,薛惠锋,李宁, 等. 基于贝叶斯神经网络遗传算法的锅炉燃烧优化[J]. 系统仿真学报, 2015, 27(8): 1790-1795.FANG Haiquan, XUE Huifeng, LI Ning, et al. Boiler combustion optimization based on Bayesian neural network and genetic algorithm[J]. Journal of System Simulation, 2015, 27(8): 1790-1795.

沈奇. 锅炉燃烧动态建模及优化控制方法研究[D]. 南京: 东南大学, 2017.

CUOMO S, DI COLA V S,GIAMPAOLO F, et al. Scientific machine learning through physics-informed neural networks: where we are and what's next[J]. Journal of Scientific Computing, 2022, 92(3): 88.

KARNIADAKIS G E, KEVREKIDIS I G, LU Lu, et al. Physics-informed machine learning[J]. Nature Reviews Physics, 2021, 3: 421-434.

任少君, 朱保宇, 翁琪航, 等. 基于物理信息神经网络的燃煤锅炉NOx排放浓度预测方法[J]. 中国电机工程学报, 2024, 44(20): 8157-8165.REN Shaojun, ZHU Baoyu, WENG Qihang, et al. Forecasting method for NOx emission in coal fired boiler based on physics-informed neural network[J]. Proceedings of the CSEE, 2024, 44(20): 8157-8165.

ARZANI A, CASSEL K W, D' SOUZA R M. Theory-guided physics-informed neural networks for boundary layer problems with singular perturbation[J]. Journal of Computational Physics, 2023, 473: 111768.

SUYKENS J,VANDEWALLE J. Least Squares Support Vector Machine Classifiers[J]. Neural Processing Letters,1999,9(3):293-300.

YANG Xiaowei,LU Jie,et al. Adaptive pruning algorithm for least squares support vector machine classifier[J]. Soft Computing,2010,14:667-680.

郭文康, 莫正阳, 李益国. 基于改进自适应最小二乘支持向量机的飞灰含碳量软测量方法[J]. 动力工程学报. 2025, 45(7): 1082-1090.GUO Wenkang, MO Zhengyang, LI Yiguo. A soft measurement method of fly ash carbon content based on improved adaptive least squares support vector machine[J]. Journal of Chinese Society of Power Engineering. 2025, 45(7): 1082-1090.

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

风功率预测关键技术及其研究应用综述

重型燃气轮机动叶片蠕变寿命预测的轻量化建模方法

基于深度学习的锅炉热效率动态预测模型研究

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

基于物理模型与机器学习求解叶栅流场的降阶模型

基于物理单调最小二乘支持向量机的锅炉燃烧NOx生成浓度预测方法

Prediction Method of Boiler Combustion NOx Generation Concentration Based on Physics-constrained Monotonic Least Squares Support Vector Machine

DOI：10.19805/j.cnki.jcspe.2026.250044

摘要

Abstract

关键词

Keywords

references

基于物理单调最小二乘支持向量机的锅炉燃烧NO_x生成浓度预测方法

Prediction Method of Boiler Combustion NO_x Generation Concentration Based on Physics-constrained Monotonic Least Squares Support Vector Machine