图像特征与振动频谱多源融合驱动的煤矸识别技术研究  

Research on coal-gangue identification technology driven by multi-source fusion of image features and vibration spectrum

作　　者：李立宝 袁永[1,2] 秦正寒 李波[1,2] 闫政天[1,2] 李勇 LI Libao;YUAN Yong;QIN Zhenghan;LI Bo;YAN Zhengtian;LI Yong(School of Mines,China University of Miming and Technology,Xuzhou 221116,China;Key Laboratory of Deep Coal Resource,Ministry of Education of China,China University of Mining and Technology,Xuzhou 221116,China)

机构地区：[1]中国矿业大学矿业工程学院,江苏徐州221116 [2]中国矿业大学深部煤炭资源开采教育部重点实验室,江苏徐州221116

出　　处：《工矿自动化》2024年第11期43-51,共9页Journal Of Mine Automation

基　　金：国家自然科学基金项目(52204132);江苏高校“青蓝工程”资助项目(苏教师函[2022]29号);江苏省研究生科研与实践创新计划资助项目(KYCX24_2874);中国矿业大学未来杰出人才助力计划资助项目(2024WLJCRCZL013);湖南省自然科学基金青年项目(2023JJ40285);湖南省教育厅优秀青年基金项目(22B0469)。

摘　　要：针对目前图像与振动信号融合的方法在煤矸识别领域应用存在特征融合困难、实时性和模型复杂度不满足实际应用要求等问题,设计了基于多头注意力(MA)的多层长短期记忆(ML-LSTM)模型MA-ML-LSTM。采用经粒子群优化(PSO)算法优化的变分模态分解(VMD)算法对振动信号进行处理,将能量、能量矩、峭度、波形因数与矩阵奇异值作为特征量,并采用一维卷积网络获取振动信息;在多分类网络ResNet-18基础上删除最后的全连接层,用于对煤矸图像进行深度特征提取;通过MA机制和ML-LSTM网络实现图像与振动双通道特征融合,强化各通道重要特征信息的表达。实验结果表明:MA-ML-LSTM模型的平均识别准确率达98.72%,相比传统单一的ResNet,MobilenetV3,1D-CNN,LSTM模型分别高4.60%,7.96%,5.37%,6.11%,相比EMD-RF,IMF-SVM,CSPNet-YOLOv7分别高4.18%,4.45%,3.46%,验证了图像特征与振动频谱多源融合驱动的煤矸识别技术的有效性。To address the challenges of feature fusion,real-time performance,and model complexity in the application of image and vibration signal fusion for coal-gangue identification,a multi-head attention(MA)-based multi-layer long short-term memory(ML-LSTM)model,i.e.,MA-ML-LSTM,was proposed.The variational mode decomposition(VMD)algorithm,optimized by particle swarm optimization(PSO),was employed to process vibration signals.Features such as energy,energy moment,kurtosis,waveform factor,and matrix singular values were extracted.A one-dimensional convolutional network was used to acquire vibration information.For image feature extraction,the fully connected layer of the multi-classification network ResNet-18 was removed,enabling the extraction of deep features from coal-gangue images.Dual-channel feature fusion of images and vibration signals was achieved using the MA mechanism and the ML-LSTM network,enhancing the expression of significant features in each channel.Experimental results demonstrated that the MA-ML-LSTM model achieved an average recognition accuracy of 98.72%,which was 4.60%,7.96%,5.37%,and 6.11%higher than traditional single models ResNet,MobilenetV3,1D-CNN,and LSTM,respectively.Compared to EMD-RF,IMF-SVM,and CSPNet-YOLOv7 models,accuracy improved by 4.18%,4.45%,and 3.46%,respectively.These findings validate the effectiveness of the coal-gangue identification technology driven by multi-source fusion of image features and vibration spectrum.

关 键 词：煤矸识别 多源信息融合 振动信号 图像识别 多头注意力机制 多层长短期记忆模型 

分 类 号：TD823.49[矿业工程—煤矿开采]