基于特征选择与BiLSTM多变量回归预测的磨煤机故障预警研究

罗云; 李战国; 付陇霞; 王道谊; 张新中; 李耀华; 程亮; 江霞

doi:10.19805/j.cnki.jcspe.2025.240145

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 >

基于特征选择与BiLSTM多变量回归预测的磨煤机故障预警研究

数字化与智能化 | 更新时间：2025-10-16

- 基于特征选择与BiLSTM多变量回归预测的磨煤机故障预警研究
- Research on Fault Warning of Coal Mill Based on Feature Selection and Multivariate Regression Prediction with BiLSTM
- 动力工程学报 2025年45卷第5期页码：724-732
- 作者机构：
  
  1. 北京大唐高鸿数据网络技术有限公司,北京,100195
  2. 华北电力科学研究院有限责任公司,北京,100045
- 作者简介：
  
  [ "罗云(1981—),男,湖南衡阳人,高级工程师,本科,研究方向为大数据建模和智慧电厂应用" ]
- 基金信息：
- DOI：10.19805/j.cnki.jcspe.2025.240145
  中图分类号： TP18;TM621
- 网络出版：2025-05-19，
  
  纸质出版：2025
- 稿件说明：
移动端阅览
罗云,李战国,付陇霞,王道谊,张新中,李耀华,程亮,江霞. 基于特征选择与BiLSTM多变量回归预测的磨煤机故障预警研究动力工程学报, 2025, 45(5): 724-732 https://doi.

org/10.19805/j.cnki.jcspe.2025.240145
罗云,李战国,付陇霞,王道谊,张新中,李耀华,程亮,江霞. 基于特征选择与BiLSTM多变量回归预测的磨煤机故障预警研究动力工程学报, 2025, 45(5): 724-732 https://doi. DOI： 10.19805/j.cnki.jcspe.2025.240145.

org/10.19805/j.cnki.jcspe.2025.240145 DOI：

摘要

为解决火电设备多参数耦合、渐变性故障诊断困难的问题

提出了一种基于套索(LASSO) 回归特征选择与双向长短期记忆(BiLSTM)网络多变量回归预测的故障预警方法。以某1 000 MW机组磨煤机为研究对象

选取磨煤机电流、出口压力、出入口差压作为表征堵磨故障的特征参数

采用LASSO回归选择特征变量

基于BiLSTM算法建立多变量回归预测模型;根据堵磨时特征参数的变化机理与模型预测值构建堵磨故障指数

最后利用核密度估计方法计算预警阈值

实现了堵磨故障预警。通过实际数据分析表明:磨煤机正常状态时

BiLSTM多变量回归预测模型的平均相对误差为1.13%

相比传统的误差反向传播(BP)神经网络和支持向量机回归(SVR)模型具有更高的精度和预测参数变化趋势的能力;磨煤机异常状态时

相比成熟的多元状态估计技术(MSET)算法模型能更早地发现磨煤机运行的异常状态

实现磨煤机变工况下故障早期预警。

Abstract

To address the difficulty of fault diagnosis in thermal power equipment with multi-parameter coupling and gradual changes

a fault early warning method based on least absolute shrinkage and selection operator (LASSO) regression feature selection and bidirectional long-short term memory (BiLSTM) multivariate regression prediction was proposed. Taking a coal mill in a 1 000 MW power unit as the research subject

feature parameters such as mill current

outlet pressure

and inlet-outlet differential pressure were selected to represent blockage faults. LASSO regression was employed to select the feature variables

and a multivariate regression prediction model was established based on the BiLSTM algorithm. According to the variation mechanism of the feature parameters during mill blockage and the predicted values of the model

a mill blockage fault index was constructed. Finally

the warning threshold was determined using the kernel density estimation method

enabling mill blockage fault warnings. Actual data analysis shows that when the coal mill is operating normally

the average relative error of the BiLSTM multivariate regression prediction model is 1.13%. Compared with the traditional error back-propagation (BP) neural network and support vector regression (SVR) model

it has higher accuracy and the ability to predict the trend of parameter change. When the coal mill is operating abnormally

this method can detect operational abnormalities earlier than the multivariate state estimation technique (MSET) algorithm model

enabling early fault warning under variable operating conditions of the coal mill.

关键词

Keywords

references

章建华, 林山青, 余兵, 等. 新型电力系统发展蓝皮书[R]. 北京: 中国电力出版社, 2023.

杨婷婷, 高乾, 李浩千, 等. 基于卷积神经网络-长短时记忆神经网络的磨煤机故障预警[J]. 热力发电, 2022, 51(10): 122-129. YANG Tingting, GAO Qian, LI Haoqian, et al. Coal mill fault early warning technology based on CNN-LSTM network[J]. Thermal Power Generation, 2022, 51(10): 122-129.

李晓彬. 低负荷下电站辅机安全监控技术研究与应用[D]. 北京: 华北电力大学, 2020.

杨婷婷, 张蓓, 吕游, 等. 基于MSET的电站风机故障预警技术研究[J]. 热能动力工程, 2017, 32(9): 63-68. YANG Tingting, ZHANG Bei, LV You, et al. Research on early fault warning technology of fan in power plant based on MSET method[J]. Journal of Engineering for Thermal Energy and Power, 2017, 32(9): 63-68.

蔡吉磊, 茅大钧, 李玉珍. 基于改进HAC和MSET的给水泵故障预警方法[J]. 热能动力工程, 2023, 38(8): 176-184. CAI Jilei, MAO Dajun, LI Yuzhen. Feed water pump fault early warning method based on improved HAC and MSET[J]. Journal of Engineering for Thermal Energy and Power, 2023, 38(8): 176-184.

韩万里, 茅大钧, 印琪民. 基于PCA和多元状态估计的引风机故障预警[J]. 热能动力工程, 2020, 35(1): 91-97. HAN Wanli, MAO Dajun, YIN Qimin. Induced draft fan fault warning based on PCA and multivariate state estimation technique[J]. Journal of Engineering for Thermal Energy and Power, 2020, 35(1): 91-97.

宋莹, 张钟镭. 基于多元状态估计和超椭球分析的汽轮机运行状态监测[J]. 动力工程学报, 2020, 40(2): 138-144. SONG Ying, ZHANG Zhonglei. Steam turbine condition monitoring based on multivariate state estimation technique and hyper-ellipsoid analysis[J]. Journal of Chinese Society of Power Engineering, 2020, 40(2): 138-144.

牛玉广, 李晓彬, 张佳辉. 基于多元状态估计与自适应阈值的电站辅机故障预警[J]. 动力工程学报, 2019, 39(9): 717-724. NIU Yuguang, LI Xiaobin, ZHANG Jiahui. Early fault warning for auxiliary equipment of power plants based on multivariate state estimation and adaptive threshold[J]. Journal of Chinese Society of Power Engineering, 2019, 39(9): 717-724.

刘海涛. 中速磨煤机状态监测与故障诊断方法研究[D]. 北京: 华北电力大学, 2015.

杨茜. 基于数据驱动的火电厂关键辅机设备故障诊断技术研究[D]. 杭州: 浙江大学, 2017.

陈波, 王亚欧, 陶谦, 等. 基于长短期记忆神经网络的燃煤机组磨煤机性能预测模型[J]. 热力发电, 2021, 50(6): 121-127. CHEN Bo, WANG Yaou, TAO Qian, et al. LSTM-based performance prediction model for coal mill in a coal-fired unit[J]. Thermal Power Generation, 2021, 50(6): 121-127.

安吉振, 乔世超, 陈衡, 等. 基于多元状态估计和向量相似度的电站磨煤机故障智能预警研究[J]. 热力发电, 2022, 51(12): 64-71. AN Jizhen, QIAO Shichao, CHEN Heng, et al. Research on intelligent early warning of coal mill failure in power station based on multi-state estimation and vector similarity[J]. Thermal Power Generation, 2022, 51(12): 64-71.

TIBSHIRANI R. Regression shrinkage and selection via the Lasso[J]. Journal of the Royal Statistical Society: Series B (Methodological), 1996, 58(1): 267-288.

沈曙昀. 基于Lasso改进方法的复杂数据联合模型变量选择[D]. 武汉: 华中科技大学, 2019.

程侃如, 王玉璋, 杨志鹏, 等. 基于SSA和注意力机制BiLSTM的燃气轮机传感器故障诊断方法研究[J]. 动力工程学报, 2023, 43(9): 1181-1189. CHENG Kanru, WANG Yuzhang, YANG Zhipeng, et al. Gas turbine sensor fault diagnosis method based on SSA and BiLSTM with attention mechanism[J]. Journal of Chinese Society of Power Engineering, 2023, 43(9): 1181-1189.

金志远, 李胜男, 谭鹏, 等. 基于长短时记忆神经网络的锅炉多参数协同预测模型[J]. 热力发电, 2021, 50(5): 120-126. JIN Zhiyuan, LI Shengnan, TAN Peng, et al. Multi-parameter collaborative prediction model of boilers based on long-short-term memory neural network[J]. Thermal Power Generation, 2021, 50(5): 120-126.

HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.

向玲, 朱浩伟, 丁显, 等. 基于CAE与BiLSTM结合的风电机组齿轮箱故障预警方法研究[J]. 动力工程学报, 2022, 42(6): 514-521. XIANG Ling, ZHU Haowei, DING Xian, et al. Research on fault warning method of wind turbine gearbox based on CAE and BiLSTM[J]. Journal of Chinese Society of Power Engineering, 2022, 42(6): 514-521.

李麒,朱光平. 基于 SWT 和 SVR 的重力坝变形预测研究[J]. 人民长江, 2021, 52(11): 169-174. LI Qi, ZHU Guangping. Deformation prediction of gravity dams based on SWT and SVR[J]. Yangtze River, 2021, 52(11): 169-174.

李昊, 罗云, 李瑞东, 等. 基于粒子群优化算法优化BP神经网络模型的间接空冷散热器性能监测[J]. 动力工程学报, 2019, 39(12): 973-980. LI Hao, LUO Yun, LI Ruidong, et al. Performance monitoring of indirect air-cooled radiators based on particle swarm optimization BP neural network model[J]. Journal of Chinese Society of Power Engineering, 2019, 39(12): 973-980.

吴涛, 王毅, 刘震, 等. 基于多元状态估计-层次分析法的磨煤机故障预警[J]. 热力发电, 2023, 52(5): 14-21. WU Tao, WANG Yi, LIU Zhen, et al. Coal mill fault warning based on multi-state estimation-analytic hierarchy process[J]. Thermal Power Generation, 2023, 52(5): 14-21.

魏书荣, 张鑫, 符杨, 等. 基于GRA-LSTM-Stacking模型的海上双馈风力发电机早期故障预警与诊断[J]. 中国电机工程学报, 2021, 41(7): 2373-2382. WEI Shurong, ZHANG Xin, FU Yang, et al. Early fault warning and diagnosis of offshore wind DFIG based on GRA-LSTM-Stacking model[J]. Proceedings of the CSEE, 2021, 41(7): 2373-2382.

杨婷婷, 李浩千, 陈晓峰, 等. 基于WPT-Transformer的磨煤机故障预警研究[J]. 热力发电, 2023, 52(12): 180-189. YANG Tingting, LI Haoqian, CHEN Xiaofeng, et al. Research on fault early warning of coal mill based on WPT and Transformer[J]. Thermal Power Generation, 2023, 52(12): 180-189.

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

直吹式制粉系统风粉偏差及可调缩孔调平的数值模拟研究

基于时空图注意网络的磨煤机故障预警