View in EDS PDF Full Text

基于频率动态感知与多视角树形拓扑的 掘进机截割头故障诊断.

Saved in:
Bibliographic Details
Title: 基于频率动态感知与多视角树形拓扑的 掘进机截割头故障诊断.
Alternate Title: Fault diagnosis of roadheader cutting heads based on frequency dynamic-aware and multi-view tree-structured topology.
Authors: 苏树智1 sushuzhi@foxmail.com, 杨梦洋1, 马天兵2 dfmtb@163.com, 朱彦敏2,3, 王 贤1
Source: Coal Science & Technology (0253-2336). May2026, Vol. 54 Issue 5, p178-194. 17p.
Subject Terms: *Fault diagnosis, *Feature extraction, *Convolutional neural networks, *Noise control
Abstract (English): As the primary equipment for roadway excavation, the roadheader operates under highly complex and variable working conditions, which makes the cutting head prone to faults. However, harsh excavation environments cause the cutting signals of the roadheader to exhibit strong background noise coupling and pronounced non-stationary characteristics, which significantly hinder fault feature extraction and lead to low diagnostic accuracy. To address these challenges, a Frequency Dynamic-aware Multi-view Tree-topology Diagnosis Network (FDM-TreeDNet) is proposed. In this network, a novel Frequency Dynamic-Aware Convolution (FDAC) module is designed. The module exploits the spectral centroid of channels to guide channel reordering and, by leveraging the energy distribution characteristics of high- and low-frequency signals, employs anisotropic convolutions to accurately decouple high- and low-frequency features. Subsequently, a multi-branch convolutional gating mechanism is adopted to capture real-time semantic information and dynamically generate samplespecific weights. While achieving frequency awareness, the FDAC module also performs adaptive feature reconstruction and calibration, thereby suppressing strong background noise and effectively alleviating feature drift under non-stationary operating conditions with severe noise interference.To further address the issue of semantic degradation during deep feature propagation, a Source-Guided Multi-view Treetopology (SGM-Tree) structure is constructed. In this structure, lateral injection paths are established at key stages of the backbone network, where parallel branches explicitly decompose multi-view critical features from the raw signal, including precise positional information, long-range structural representations, and global contextual semantics. These features are progressively injected as semantic anchors prior to the FDAC modules, effectively correcting semantic deviations introduced by frequency reordering. To fully exploit the hierarchical feature advantages brought by the tree-topology architecture, a Tree-Structured Distillation (TSD) strategy is further introduced. Through a parameter-sharing classification mechanism, the multi-view critical features injected by the SGM-Tree at different stages and the dynamic output features of the backbone network are projected into a unified semantic metric space. Within this space, confidence transformation is employed to obtain semantic soft masks, which, combined with multi-level cascaded self-distillation, enhance the fault diagnosis accuracy of the roadheader cutting head. Experimental results demonstrate that the proposed roadheader cutting head fault diagnosis approach achieves superior performance on both the self-constructed AUST roadheader dataset and the CWRU dataset, thereby validating its effectiveness. [ABSTRACT FROM AUTHOR]
Abstract (Chinese): 掘进机作为巷道掘进的主力装备, 其掘进工况却复杂多变, 极易导致掘进机截割头发生故障。 恶劣的掘进工况, 导致掘进机截割信号通常存在强背景噪声耦合与高度非平稳特性的问题, 这使得 故障特征提取困难, 故障诊断准确率低。针对这一问题, 提出一种频率动态感知的多视角树形拓扑 诊断网络 (Frequency Dynamic-aware Multi-view Tree-topology Diagnosis Network, FDM-TreeDNet)。 该网络设计了一种新颖的频率动态感知卷积 (Frequency Dynamic-Aware Convolution, FDAC) 模块, 该模块利用通道的频谱质心指导通道重排, 并结合高低频信号的能量分布特性, 使用各向异性卷积 对高低频特征精准解耦, 随后经过多支路卷积门控机制, 捕捉实时语义来动态生成样本特异性权重。 FDAC 在实现频率感知的同时, 也对特征图进行了自适应重构与校准, 从而抑制强背景噪声, 有效 克服了非平稳工况与强背景噪声下的特征漂移困境。为了进一步解决该模块在深层特征传递中的语 义迷失问题, 设计了源流引导的多视角树形拓扑 (Source-Guided Multi-view Tree-topology, SGM-Tree) 结构, 该结构在主干网络的各个关键阶段建立横向注入通道, 利用并行支路从原始信号中显式分解 出精确坐标、长程结构及全局上下文等多视角关键特征。这些特征作为语义锚点被逐阶段注入到 FDAC 模块之前, 有效校准了因频率重排引起的语义偏差。为了充分挖掘树形拓扑结构带来的层级 化特征优势, 进一步提出了一种面向树形结构的蒸馏 (Tree-Structured Distillation, TSD) 策略, 该策 略利用参数共享分类机制将各阶段 SGM-Tree 注入的多视角关键特征与主干网络的动态输出特征映 射至同一语义度量空间, 并在该空间内借助置信变换来获取语义软掩码, 结合多层联级自蒸馏, 提 高掘进机截割头故障诊断准确率。试验表明, 该掘进机截割头故障诊断方法在自建 AUST 掘进机数 据集和 CWRU 数据集上效果显著, 验证了其有效性。 [ABSTRACT FROM AUTHOR]
Database: Energy & Power Source
FullText Links:
  – Type: pdflink
Text:
  Availability: 0
Header DbId: enr
DbLabel: Energy & Power Source
An: 194562349
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: 基于频率动态感知与多视角树形拓扑的 掘进机截割头故障诊断.
– Name: TitleAlt
  Label: Alternate Title
  Group: TiAlt
  Data: Fault diagnosis of roadheader cutting heads based on frequency dynamic-aware and multi-view tree-structured topology.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22苏树智%22">苏树智</searchLink><relatesTo>1</relatesTo><i> sushuzhi@foxmail.com</i><br /><searchLink fieldCode="AR" term="%22杨梦洋%22">杨梦洋</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22马天兵%22">马天兵</searchLink><relatesTo>2</relatesTo><i> dfmtb@163.com</i><br /><searchLink fieldCode="AR" term="%22朱彦敏%22">朱彦敏</searchLink><relatesTo>2,3</relatesTo><br /><searchLink fieldCode="AR" term="%22王+贤%22">王 贤</searchLink><relatesTo>1</relatesTo>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Coal+Science+%26+Technology+%280253-2336%29%22">Coal Science & Technology (0253-2336)</searchLink>. May2026, Vol. 54 Issue 5, p178-194. 17p.
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: *<searchLink fieldCode="DE" term="%22Fault+diagnosis%22">Fault diagnosis</searchLink><br />*<searchLink fieldCode="DE" term="%22Feature+extraction%22">Feature extraction</searchLink><br />*<searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink><br />*<searchLink fieldCode="DE" term="%22Noise+control%22">Noise control</searchLink>
– Name: Abstract
  Label: Abstract (English)
  Group: Ab
  Data: As the primary equipment for roadway excavation, the roadheader operates under highly complex and variable working conditions, which makes the cutting head prone to faults. However, harsh excavation environments cause the cutting signals of the roadheader to exhibit strong background noise coupling and pronounced non-stationary characteristics, which significantly hinder fault feature extraction and lead to low diagnostic accuracy. To address these challenges, a Frequency Dynamic-aware Multi-view Tree-topology Diagnosis Network (FDM-TreeDNet) is proposed. In this network, a novel Frequency Dynamic-Aware Convolution (FDAC) module is designed. The module exploits the spectral centroid of channels to guide channel reordering and, by leveraging the energy distribution characteristics of high- and low-frequency signals, employs anisotropic convolutions to accurately decouple high- and low-frequency features. Subsequently, a multi-branch convolutional gating mechanism is adopted to capture real-time semantic information and dynamically generate samplespecific weights. While achieving frequency awareness, the FDAC module also performs adaptive feature reconstruction and calibration, thereby suppressing strong background noise and effectively alleviating feature drift under non-stationary operating conditions with severe noise interference.To further address the issue of semantic degradation during deep feature propagation, a Source-Guided Multi-view Treetopology (SGM-Tree) structure is constructed. In this structure, lateral injection paths are established at key stages of the backbone network, where parallel branches explicitly decompose multi-view critical features from the raw signal, including precise positional information, long-range structural representations, and global contextual semantics. These features are progressively injected as semantic anchors prior to the FDAC modules, effectively correcting semantic deviations introduced by frequency reordering. To fully exploit the hierarchical feature advantages brought by the tree-topology architecture, a Tree-Structured Distillation (TSD) strategy is further introduced. Through a parameter-sharing classification mechanism, the multi-view critical features injected by the SGM-Tree at different stages and the dynamic output features of the backbone network are projected into a unified semantic metric space. Within this space, confidence transformation is employed to obtain semantic soft masks, which, combined with multi-level cascaded self-distillation, enhance the fault diagnosis accuracy of the roadheader cutting head. Experimental results demonstrate that the proposed roadheader cutting head fault diagnosis approach achieves superior performance on both the self-constructed AUST roadheader dataset and the CWRU dataset, thereby validating its effectiveness. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label: Abstract (Chinese)
  Group: Ab
  Data: 掘进机作为巷道掘进的主力装备, 其掘进工况却复杂多变, 极易导致掘进机截割头发生故障。 恶劣的掘进工况, 导致掘进机截割信号通常存在强背景噪声耦合与高度非平稳特性的问题, 这使得 故障特征提取困难, 故障诊断准确率低。针对这一问题, 提出一种频率动态感知的多视角树形拓扑 诊断网络 (Frequency Dynamic-aware Multi-view Tree-topology Diagnosis Network, FDM-TreeDNet)。 该网络设计了一种新颖的频率动态感知卷积 (Frequency Dynamic-Aware Convolution, FDAC) 模块, 该模块利用通道的频谱质心指导通道重排, 并结合高低频信号的能量分布特性, 使用各向异性卷积 对高低频特征精准解耦, 随后经过多支路卷积门控机制, 捕捉实时语义来动态生成样本特异性权重。 FDAC 在实现频率感知的同时, 也对特征图进行了自适应重构与校准, 从而抑制强背景噪声, 有效 克服了非平稳工况与强背景噪声下的特征漂移困境。为了进一步解决该模块在深层特征传递中的语 义迷失问题, 设计了源流引导的多视角树形拓扑 (Source-Guided Multi-view Tree-topology, SGM-Tree) 结构, 该结构在主干网络的各个关键阶段建立横向注入通道, 利用并行支路从原始信号中显式分解 出精确坐标、长程结构及全局上下文等多视角关键特征。这些特征作为语义锚点被逐阶段注入到 FDAC 模块之前, 有效校准了因频率重排引起的语义偏差。为了充分挖掘树形拓扑结构带来的层级 化特征优势, 进一步提出了一种面向树形结构的蒸馏 (Tree-Structured Distillation, TSD) 策略, 该策 略利用参数共享分类机制将各阶段 SGM-Tree 注入的多视角关键特征与主干网络的动态输出特征映 射至同一语义度量空间, 并在该空间内借助置信变换来获取语义软掩码, 结合多层联级自蒸馏, 提 高掘进机截割头故障诊断准确率。试验表明, 该掘进机截割头故障诊断方法在自建 AUST 掘进机数 据集和 CWRU 数据集上效果显著, 验证了其有效性。 [ABSTRACT FROM AUTHOR]
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=194562349
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.12438/cst.2025−1892
    Languages:
      – Code: chi
        Text: Chinese
    PhysicalDescription:
      Pagination:
        PageCount: 17
        StartPage: 178
    Subjects:
      – SubjectFull: Fault diagnosis
        Type: general
      – SubjectFull: Feature extraction
        Type: general
      – SubjectFull: Convolutional neural networks
        Type: general
      – SubjectFull: Noise control
        Type: general
    Titles:
      – TitleFull: 基于频率动态感知与多视角树形拓扑的 掘进机截割头故障诊断.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: 苏树智
      – PersonEntity:
          Name:
            NameFull: 杨梦洋
      – PersonEntity:
          Name:
            NameFull: 马天兵
      – PersonEntity:
          Name:
            NameFull: 朱彦敏
      – PersonEntity:
          Name:
            NameFull: 王 贤
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 02532336
          Numbering:
            – Type: volume
              Value: 54
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: Coal Science & Technology (0253-2336)
              Type: main
ResultId 1