Intelligent recognition of coal and rock based on wavelet packet multi-scale fuzzy entropy and weighted KL divergence
-
摘要: 垮落煤岩智能识别是智能放煤的前提，通过垮落煤岩实时精准识别可避免人工放煤造成的顶煤“欠放”或“过放”问题。现有煤岩识别方法大多通过数据降维处理获得垮落煤岩特征向量，通过构建识别模型进行煤岩识别，但数据降维、模型建立和训练均需较长时间，一定程度上影响了连续综放开采效率。针对该问题，提出了一种基于小波包多尺度模糊熵和加权KL散度的煤岩智能识别方法。对不同工况（顶煤垮落、岩石垮落、大块顶煤垮落）下垮落煤岩冲击液压支架后尾梁的振动信号进行小波包分解，得到一系列频带；对各频带的序列进行粗粒化，计算各频带多个尺度粗粒化向量的模糊熵，即小波包多尺度模糊熵，将其作为特征向量；以小波包分解后各频带能量与振动信号总能量的比值作为加权KL散度的权重，比较待测未知样本与不同工况下样本特征向量的加权KL散度，实现垮落煤岩的实时精准识别。实验结果表明：基于小波包多尺度模糊熵和加权KL散度的方法可有效识别垮落煤岩类别，而基于多尺度模糊熵和KL散度的方法、基于小波包模糊熵和KL散度的方法识别效果不佳；将小波包多尺度模糊熵作为特征向量时，BP神经网络识别准确率达95%，进一步验证了小波包多尺度模糊熵可作为表征垮落煤岩的特征向量；整个煤岩识别过程耗时为1.063 9 s，基本满足垮落煤岩智能识别实时性需求，大大降低了对连续综放开采效率的影响，综合性能优于同类煤岩识别方法。Abstract: Intelligent recognition of collapsed coal and rock is a prerequisite for intelligent coal caving. Real-time and precise recognition of collapsed coal and rock can avoid the problem of "under caving" or "over caving" of top coal caused by manual coal caving. Most existing coal and rock recognition methods obtain collapsed coal and rock feature vectors through data dimensionality reduction processing, and construct recognition models for coal and rock recognition. However, data dimensionality reduction, model establishment, and training all require a long time. To some extent, these factors affect the efficiency of continuous fully mechanized caving mining. In order to solve the above problems, an intelligent coal and rock recognition method based on wavelet packet multi-scale fuzzy entropy and weighted KL divergence is proposed. Wavelet packet decomposition is performed on the vibration signals of the tail beam after the hydraulic support is impacted by collapsed coal and rock under different working conditions (top coal collapse, rock collapse, and large top coal collapse) to obtain a series of frequency bands. The sequences of each frequency band are coarse-grained. The method calculates the fuzzy entropy under multiple scales of coarse-grained sequences in each frequency band, that is, wavelet packet multi-scale fuzzy entropy. The method uses it as a feature vector. The method uses the ratio of the energy of each frequency band after wavelet packet decomposition to the total energy of the vibration signal as the weight of the weighted KL divergence. The weighted KL divergence of the unknown samples to be tested and the sample feature vectors under different working conditions are compared. The real-time and precise recognition of collapsed coal and rock is achieved. The experimental results show that the method based on wavelet packet multi-scale fuzzy entropy and weighted KL divergence can effectively recognize the category of collapsed coal and rock. The method based on multi-scale fuzzy entropy and KL divergence and the method based on wavelet packet fuzzy entropy and KL divergence have poor recognition performance. When wavelet packet multi-scale fuzzy entropy is used as the feature vector, the recognition accuracy of the BP neural network reaches 95%. It further verifies that wavelet packet multi-scale fuzzy entropy can be used as the feature vector to characterize collapsed coal and rock. The entire coal and rock identification process takes 1.063 9 seconds, which basically meets the real-time requirements of intelligent recognition of collapsed coal and rock. At the same time, it greatly reduces the impact on the efficiency of continuous fully mechanized caving mining. Its comprehensive performance is superior to similar coal and rock recognition methods.
-
图 3 不同工况下振动信号时域波形
Figure 3. Time domain waveform of vibration signals under different working conditions
图 4 不同样本小波包分解后各频带重构信号
Figure 4. Reconstructed signals in different frequency bands after wavelet packet decomposition of different samples
图 5 不同方法煤岩识别结果对比
Figure 5. Comparison of coal and rock recognition results using different methods
表 1 BP神经网络训练和识别结果对比
Table 1. Comparison of BP neural network training and recognition results
下载: 导出CSV
表 2 垮落煤岩识别方法性能对比
Table 2. Performance comparison of recognition methods for collapsed coal and rock
识别方法 耗时/s 识别准确率/% 小波包多尺度模糊熵+加权KL散度 1.063 9 95 （EMD）IMF能量+BP神经网络 17.031 7 85 分形盒维数和小波包能量矩+ BP神经网络 18.223 1 95 EEMD+KPCA+KL散度 1.741 6 95
下载: 导出CSV
-
[1] 李一鸣,符世琛,李瑞,等. 垮落煤岩性状识别研究[J]. 工矿自动化,2017,43(2):24-28. doi: 10.13272/j.issn.1671-251x.2017.02.006LI Yiming,FU Shichen,LI Rui,et al. Reseaech on identification of caving coal and rock traits[J]. Industry and Mine Automation,2017,43(2):24-28. doi: 10.13272/j.issn.1671-251x.2017.02.006 [2] 吕渊博,王世博,葛世荣,等. 近红外光谱煤岩识别装置研制[J]. 工矿自动化,2022,48(7):32-42. doi: 10.13272/j.issn.1671-251x.17953LYU Yuanbo,WANG Shibo,GE Shirong,et al. Development of coal and rock identification device based on near-infrared spectroscopy[J]. Journal of Mine Automation,2022,48(7):32-42. doi: 10.13272/j.issn.1671-251x.17953 [3] 卢召栋. 近红外光谱煤岩识别算法及风速影响研究[D]. 徐州: 中国矿业大学, 2022.LU Zhaodong. Study on algorithm of near infrared spectroscopy for coal and rock identification and influence of wind velocity on it[D]. Xuzhou: China University of Mining and Technology, 2022. [4] 毛青晨. 基于图像处理的煤岩智能识别方法研究[D]. 徐州: 中国矿业大学, 2022.MAO Qingchen. Research on intelligent identification method of coal and rock based on image processing[D]. Xuzhou: China University of Mining and Technology, 2022. [5] 高峰,殷欣,刘泉声,等. 基于塔式池化架构的采掘工作面煤岩图像识别方法[J]. 煤炭学报,2021,46(12):4088-4102. doi: 10.13225/j.cnki.jccs.2021.0624GAO Feng,YIN Xin,LIU Quansheng,et al. Coal-rock image recognition method for mining and heading face based on spatial pyramid pooling structure[J]. Journal of China Coal Society,2021,46(12):4088-4102. doi: 10.13225/j.cnki.jccs.2021.0624 [6] 张斌,苏学贵,段振雄,等. YOLOv2在煤岩智能识别与定位中的应用研究[J]. 采矿与岩层控制工程学报,2020,2(2):94-101.ZHANG Bin,SU Xuegui,DUAN Zhenxiong,et al. Application of YOLOv2 in intelligent recognition and location of coal and rock[J]. Journal of Mining and Strata Control Engineering,2020,2(2):94-101. [7] 刘斌. 综采面煤岩自然γ射线辐射规律及识别实验研究[D]. 徐州: 中国矿业大学, 2022.LIU Bin. Experimental study on natural gamma radiation law and identification of coal and rock in fully mechanized mining face[D]. Xuzhou: China University of Mining and Technology, 2022. [8] 韦任,徐良骥,孟雪莹,等. 基于高光谱特征吸收峰的煤岩识别方法[J]. 光谱学与光谱分析,2021,41(6):1942-1948.WEI Ren,XU Liangji,MENG Xueying,et al. Coal and rock identification method based on hyper spectral feature absorption peak[J]. Spectroscopy and Spectral Analysis,2021,41(6):1942-1948. [9] 李长鹏. 采煤机声信号数据驱动截割模式识别方法研究[D]. 淮南: 安徽理工大学, 2022.LI Changpeng. Research on data-driven cutting pattern recognition method of shearer sound signal[D]. Huainan: Anhui University of Science and Technology, 2022. [10] 朱世刚. 综放工作面煤岩性状识别方法研究[D]. 北京: 中国矿业大学（北京）, 2014.ZHU Shigang. Study on coal and rock character recognition method in fully mechanized caving faces[D]. Beijing: China University of Mining & Technology-Beijing, 2014. [11] 李一鸣,符世琛,焦亚博,等. 基于分形盒维数和小波包能量矩的垮落煤岩性状识别[J]. 煤炭学报,2017,42(3):803-808. doi: 10.13225/j.cnki.jccs.2016.0729LI Yiming,FU Shichen,JIAO Yabo,et al. Collapsing coal-rock identification based on fractal box dimension and wavelet packet energy moment[J]. Journal of China Coal Society,2017,42(3):803-808. doi: 10.13225/j.cnki.jccs.2016.0729 [12] 李一鸣,符世琛,周俊莹,等. 基于小波包熵和流形学习的垮落煤岩识别[J]. 煤炭学报,2017,42(增刊2):585-593. doi: 10.13225/j.cnki.jccs.2017.0642LI Yiming,FU Shichen,ZHOU Junying,et al. Collapsing coal-rock identification based on wavelet packet entropy and manifold learning[J]. Journal of China Coal Society,2017,42(S2):585-593. doi: 10.13225/j.cnki.jccs.2017.0642 [13] 李一鸣,白龙,蒋周翔,等. 基于EEMD−KPCA和KL散度的垮落煤岩识别[J]. 煤炭学报,2020,45(2):827-835. doi: 10.13225/j.cnki.jccs.2019.0198LI Yiming,BAI Long,JIANG Zhouxiang,et al. Caving coal-rock identification based on EEMD-KPCA and KL divergence[J]. Journal of China Coal Society,2020,45(2):827-835. doi: 10.13225/j.cnki.jccs.2019.0198 [14] 刘伟,华臻,王汝琳. 基于Hilbert谱信息熵的煤矸放落振动特征分析[J]. 中国安全科学学报,2011,21(4):32-37. doi: 10.3969/j.issn.1003-3033.2011.04.006LIU Wei,HUA Zhen,WANG Rulin. Vibrational feature analysis for coal gangue caving based on information entropy of Hilbert spectrum[J]. China Safety Science Journal,2011,21(4):32-37. doi: 10.3969/j.issn.1003-3033.2011.04.006 [15] 王保平,王增才,张万枝. 基于EMD与神经网络的煤岩界面识别方法[J]. 振动. 测试与诊断,2012,32(4):586-590,688.WANG Baoping,WANG Zengcai,ZHANG Wanzhi. Coal-rock interface recognition method based on EMD and neural network[J]. Journal of Vibration,Measurement & Diagnosis,2012,32(4):586-590,688. [16] 丛晓妍,王增才,王保平,等. 基于EMD与峭度滤波的煤岩界面识别[J]. 振动. 测试与诊断,2015,35(5):950-954,996.CONG Xiaoyan,WANG Zengcai,WANG Baoping,et al. Application of filtering method based on EMD and kurtosis in coal-rock interface recognition[J]. Journal of Vibration,Measurement & Diagnosis,2015,35(5):950-954,996. [17] 郑近德,陈敏均,程军圣,等. 多尺度模糊熵及其在滚动轴承故障诊断中的应用[J]. 振动工程学报,2014,27(1):145-151. doi: 10.3969/j.issn.1004-4523.2014.01.020ZHENG Jinde,CHEN Minjun,CHENG Junsheng,et al. Multiscale fuzzy entropy and its application in rolling bearing fault diagnosis[J]. Journal of Vibration Engineering,2014,27(1):145-151. doi: 10.3969/j.issn.1004-4523.2014.01.020 [18] 李晓艳,张子刚,张逸石,等. 一种基于KL散度和类分离策略的特征选择算法[J]. 计算机科学,2012,39(12):224-227. doi: 10.3969/j.issn.1002-137X.2012.12.053LI Xiaoyan,ZHANG Zigang,ZHANG Yishi,et al. KL-divergence based feature selection algorithm with the separate-class strategy[J]. Computer Science,2012,39(12):224-227. doi: 10.3969/j.issn.1002-137X.2012.12.053 [19] 白凤兰,徐珺,刘立伟,等. 基于加权相对熵的二阶马尔可夫模型的进化树重构[J]. 大连交通大学学报,2013,34(5):112-117. doi: 10.3969/j.issn.1673-9590.2013.05.026BAI Fenglan,XU Jun,LIU Liwei,et al. Weighted relative entropy for phylogenetic tree based on 2-step Markov model[J]. Journal of Dalian Jiaotong University,2013,34(5):112-117. doi: 10.3969/j.issn.1673-9590.2013.05.026 [20] 李一鸣. 基于非平稳振动信号的综放工作面垮落煤岩识别方法研究[D]. 北京: 中国矿业大学（北京）, 2018.LI Yiming. Collapsing coal and rock recognition method based on non-stationary vibration signal in fully mechanized caving face[D]. Beijing: China University of Mining & Technology-Beijing, 2018. [21] 王保平. 放顶煤过程中煤矸界面自动识别研究[D]. 济南: 山东大学, 2012.WANG Baoping. Research on automatic recognition of the coal-rock interface in top coal caving[D]. Jinan: Shandong University, 2012. -
点击查看大图
图(5) / 表(2)
计量
- 文章访问数: 215
- HTML全文浏览量: 64
- PDF下载量: 14
- 被引次数: 0
