基于MFCC声纹识别的燃烧状态监测研究

王元; 葛沈炜; 姚鹏搏; 沈国清

doi:10.19805/j.cnki.jcspe.2025.240375

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MFCC声纹识别的燃烧状态监测研究

动力设备与系统 | 更新时间：2026-10-10

- 基于MFCC声纹识别的燃烧状态监测研究
- Study on Combustion State Monitoring Based on MFCC Voiceprint Recognition
- 动力工程学报 2025年45卷第7期页码：1010-1019
- 作者机构：
  
  1. 华北电力大学能源动力与机械工程学院,北京,102206
  2. 华北电力大学控制与计算机工程学院,北京,102206
- 作者简介：
- 基金信息：
- DOI：10.19805/j.cnki.jcspe.2025.240375
  中图分类号：
- 网络首发：2025-07-14，
  
  纸质出版：2025
- 稿件说明：
移动端阅览
王元,葛沈炜,姚鹏搏,沈国清. 基于MFCC声纹识别的燃烧状态监测研究动力工程学报, 2025, 45(7): 1010-1019 https://doi.

org/10.19805/j.cnki.jcspe.2025.240375
王元,葛沈炜,姚鹏搏,沈国清. 基于MFCC声纹识别的燃烧状态监测研究动力工程学报, 2025, 45(7): 1010-1019 https://doi. DOI： 10.19805/j.cnki.jcspe.2025.240375.

org/10.19805/j.cnki.jcspe.2025.240375 DOI：

摘要

锅炉燃烧监测对其安全运行意义重大

长期以来一直是研究的重点。燃烧产生的声音可以直接反映其状态变化

而受高温、多尘、高频噪声等因素影响

锅炉内部声音特征的有效特征提取难度极大。对此提出了一种基于一维卷积神经网络和声纹技术的燃烧特征提取方法。在构建燃烧声音采集实验平台的基础上

提取各种燃烧和非燃烧状态对应的声音特征。通过计算声纹各特征的基尼指数

筛选出火焰声纹的最佳特征。通过比较有噪声和无噪声的数据集模型

验证了火焰声纹模型的鲁棒性和准确性。结果表明:在原始数据中

声音特征识别准确率可达97.13%

优化后特征识别准确率提高到98.26%;在添加噪声的数据集中

优化后特征识别准确率可达93.29%。

Abstract

Boiler combustion monitoring is of great significance to the safe operation of boilers and has long been a focus of research. The sound generated by combustion can directly reflect its state changes. However

due to factors such as high temperature

dust

and high-frequency noise

it is extremely difficult to effectively extract the sound characteristics inside the boiler. In response to this

a combustion feature extraction method based on one-dimensional convolutional neural network and voiceprint technology was proposed. Based on the construction of a combustion sound acquisition experimental platform

the sound characteristics corresponding to various combustion and non-combustion states were extracted. By calculating the Gini index of each feature of the voiceprint

the best features of the flame voiceprint were selected. The robustness and accuracy of the flame voiceprint model were verified by comparing the noisy and noise-free dataset models. The results show that in the original data

the sound feature training accuracy can reach 97.13%

and the training accuracy after optimization is increased to 98.26%; in the dataset with added noise

the optimized feature training accuracy can reach 93.29%.

关键词

Keywords

references

赵钦新, 邓世丰, 王善武, 等. 锅炉生命周期碳减排和企业创新转型[J]. 工业锅炉, 2022(1): 1-11. ZHAO Qinxin, DENG Shifeng, WANG Shanwu, et al. Life cycle carbon emissions reduction of boilers and innovative transformation of enterprises[J]. Industrial Boilers, 2022(1): 1-11.

SU Xianqiang, FANG Qingyan, MA Lun, et al. Improving combustion and lowering NO x emissions of an industrial coal swirl burner by optimizing its nozzle structure[J ] . Applied Thermal Engineering, 2023, 218: 119340.

唐广通, 许烨烽, 闫慧博, 等. 基于深度学习与热辐射成像耦合的炉内温度场在线测量[J]. 动力工程学报, 2022, 42(10): 960-966. TANG Guangtong, XU Yefeng, YAN Huibo, et al. Research of on-line measurement of temperature field in furnaces based on deep learning coupled thermal radiative imaging[J]. Journal of Chinese Society of Power Engineering, 2022, 42(10): 960-966.

安连锁, 李庚生, 沈国清, 等. 声学测温系统在200 MW电站锅炉中的应用研究[J]. 动力工程学报, 2011, 31(12): 928-932. AN Liansuo, LI Gengsheng, SHEN Guoqing, et al. Application of acoustic pyrometer in a 200 MW power plant boiler[J]. Journal of Chinese Society of Power Engineering, 2011, 31(12): 928-932.

李旭斌, 田付强, 郭亦可. 新型电力系统中电力设备健康管理与智能运维关键技术探究[J]. 电网技术, 2023, 47(9): 3710-3726. LI Xubin, TIAN Fuqiang, GUO Yike. Key technologies for health management and intelligent operation and maintenance of power equipment in new power systems[J]. Power System Technology, 2023, 47(9): 3710-3726.

何涛, 王思睿, 卫思霁, 等. 基于深度学习的贫燃熄火预测方法和实验研究[J]. 燃烧科学与技术, 2022, 28(3): 304-312. HE Tao, WANG Sirui, WEI Siji, et al. Prediction method and experimental research on lean burn blow-off based on deep learning[J]. Journal of Combustion Science and Technology, 2022, 28(3): 304-312.

DETRICHE P, LANORE J C. An acoustic study of pulsation characteristics of fires[J]. Fire Technology, 1980, 16(3): 204-211.

YU Miao, GUPTA A, BRYDEN K M. Sensor response and sensor network development for practical combustors[C]//Proceedings of the 16th Computer Communications and Networks. Honolulu: IEEE, 2007.

SINGH A V, YU M, GUPTA A K, et al. Thermo-acoustic behavior of a swirl stabilized diffusion flame with heterogeneous sensors[J]. Applied Energy, 2013, 106: 1-16.

KHALIL A E E, GUPTA A K. Acoustic and heat release signatures for swirl assisted distributed combustion[J]. Applied Energy, 2017, 193: 125-138.

MONDAL S, BHATTACHARYA C, CHATTOPADHYAY P, et al. Prediction of thermoacoustic instabilities in a premixed combustor based on FFT-based dynamic characterization[C]//Proceedings of the 53rd AIAA/SAE/ASEE Joint Propulsion Conference. Atlanta: AIAA, 2017.

BHATTACHARYA C, DE S, MUKHOPADHYAY A, et al. Detection and classification of lean blow-out and thermoacoustic instability in turbulent combustors[J]. Applied Thermal Engineering, 2020, 180: 115808.

HANIFA R M, ISA K, MOHAMAD S. A review on speaker recognition: technology and challenges[J]. Computers & Electrical Engineering, 2021, 90: 107005.

高原, 邓艾东, 范永胜, 等. 基于MFCC和MDE-SVDD的滚动轴承音频信号异常检测方法[J]. 动力工程学报, 2024, 44(2): 277-283. GAO Yuan, DENG Aidong, FAN Yongsheng, et al. Anomaly detection method for acoustic signal of rolling bearing based on MFCC and MDE-SVDD[J]. Journal of Chinese Society of Power Engineering, 2024, 44(2): 277-283.

李楠, 马宏忠, 崔佳嘉, 等. 电力变压器绕组振动声纹特性分析[J]. 应用声学, 2022, 41(5): 785-793. LI Nan, MA Hongzhong, CUI Jiajia, et al. Characteristic analysis of vibration sound ripple of power transformer winding[J]. Journal of Applied Acoustics, 2022, 41(5): 785-793.

肖九长, 王毅, 曾小春, 等. 米勒循环汽油机燃烧噪声诊断与优化[J]. 应用声学, 2024, 43(2): 301-307. XIAO Jiuchang, WANG Yi, ZENG Xiaochun, et al. Diagnosis and optimization on combustion noise of a Miller cycle gasoline engine[J]. Journal of Applied Acoustics, 2024, 43(2): 301-307.

陈志高, 赵庆卫, 王丽, 等. 融合分布对齐和对抗学习的无监督跨域声纹识别[J]. 声学学报, 2021, 46(5): 767-774. CHEN Zhigao, ZHAO Qingwei, WANG Li, et al. Unsupervised cross-domain speaker recognition based on distribution alignment and adversarial learning[J]. Acta Acustica, 2021, 46(5): 767-774.

PICONE J W. Signal modeling techniques in speech recognition[J]. Proceedings of the IEEE, 1993, 81(9): 1215-1247.

CHOWDHURY A, ROSS A. Fusing MFCC and LPC features using 1D triplet CNN for speaker recognition in severely degraded audio signals[J]. IEEE Transactions on Information Forensics and Security, 2020, 15: 1616-1629.

KOTSIANTIS S B. Decision trees: a recent overview[J]. Artificial Intelligence Review, 2013, 39(4): 261-283.

WANG Yanzhi, WU Jinhong, YU Ziyang, et al. A structurally re-parameterized convolution neural network-based method for gearbox fault diagnosis in edge computing scenarios[J]. Engineering Applications of Artificial Intelligence, 2023, 126: 107091.

王赵国, 韦存海, 彭雅妮, 等. 基于GFCC-SVM-RFE的电力设备声音特征提取方法[J]. 电力信息与通信技术, 2022, 20(9): 34-42. WANG Zhaoguo, WEI Cunhai, PENG Yani, et al. Sound feature extraction method of power equipment based on GFCC-SVM-RFE[J]. Electric Power Information and Communication Technology, 2022, 20(9): 34-42.

王丰华, 王邵菁, 陈颂, 等. 基于改进MFCC和VQ的变压器声纹识别模型[J]. 中国电机工程学报, 2017, 37(5): 1535-1543. WANG Fenghua, WANG Shaojing, CHEN Song, et al. Voiceprint recognition model of power transformers based on improved MFCC and VQ[J]. Proceedings of the CSEE, 2017, 37(5): 1535-1543.

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

基于特征相似日和CNN-BiLSTM-Attention模型的风电短期出力预测

中心风结构对350 MW锅炉旋流燃烧器流动及燃烧特性的影响

基于CNN-BiLSTM-SSA的锅炉再热器壁温预测模型

基于WOA的循环流化床锅炉SO2排放浓度VMD-Transformer预测模型

基于不平衡样本数据的轴承故障诊断模型与方法研究