Hardness recognition method of roadheader’s cutting rock wall based on multi-source data fusion
ZHANG Linfeng1,2, TIAN Muqin1,2, SONG Jiancheng1,2, HE Ying1,2, FENG Junling1,2, LIU Xiqing3
1.Shanxi Provincial Key Lab of Mining Electrical Equipment and Intelligent Control, Taiyuan University of Technology, Taiyuan 030024, China;
2.Provincial Joint Engineering Lab of Mining Intelligent Electrical Apparatus Technology, Taiyuan University of Technology, Taiyuan 030024, China;
3.Shanxi Vocational and Technical College of Coal, Taiyuan 030024, China
Abstract：In order to solve the problem of difficulty in recognizing the hardness of cutting rock wall for roadheader in the coal mine, a method for recognizing the hardness of the cut rock wall by using the multi-source data fusion algorithm based on the cantilever vibration signal, the pressure signal of the lift cylinder and the rotary cylinder, and the current signal of the cutting motor is proposed. Firstly, the various types of signals is decomposed by wavelet packet transformation, then the signals at different frequency bands are reconstructed to construct the time-frequency matrix. The singular value decomposition is used to obtain several characteristic singular values of various signals including time-frequency information to construct feature vectors. Then the LDA algorithm is used to realize Data feature fusion, and the low-dimensional features with better class separability are obtained. In order to solve the problem that the probabilistic neural network (PNN) smoothing parameters cannot be determined and the network structure is complex, a PNN optimization method based on differential evolution algorithm (DE) and QR decomposition is proposed, and the low-dimensional features were used in hardness recognition through the processing of Optimized PNN. The experimental results show that the feature extraction method and pattern recognition method proposed in this paper are effective. And the optimized PNN has higher hardness recognition accuracy under the three working conditions of the roadheader. than other pattern recognition algorithms currently in use.
张林锋1,2，田慕琴1,2，宋建成1,2，贺颖1,2，冯君玲1,2,刘西青3. 基于多源数据融合的掘进机截割岩壁硬度识别方法[J]. 振动与冲击, 2020, 39(13): 7-15.
ZHANG Linfeng1,2, TIAN Muqin1,2, SONG Jiancheng1,2, HE Ying1,2, FENG Junling1,2, LIU Xiqing3. Hardness recognition method of roadheader’s cutting rock wall based on multi-source data fusion. JOURNAL OF VIBRATION AND SHOCK, 2020, 39(13): 7-15.
[1] 赵新利,李兰伟. 浅谈EBZ系列悬臂式掘进机的发展与应用[J]. 科技与企业, 2013, No.241(16): 285, 288.
Zhao Xinli, Li Langui. Talking about the Development and Application of EBZ Series Cantilever Roadheader[J]. Technology and enterprise, 2013, No.241(16): 285, 288.
[2] 王伟,田慕琴,宋建成,等. 基于小波包神经网络的岩巷掘进机动载荷识别方法[J]. 煤矿机械, 2015, 36(3): 238-241.
Wang Wei, LI Xiaoming, Tian Muqin, et al. Dynamic load identification method of rock roadheader based on wavelet packet and neural network[J]. Coal Mine Machinery, 2015, v.36;No.325(3): 244-247.
[3] 雷静,余斌. 基于信息融合和神经网络的煤岩识别方法[J]. 工矿自动化, 2017, 43(9): 102-105.
Lei Jing,Yu Bin. Identification method of coal and rock based on information fusion and neural network[J]. Industry and Mine Automation, 2017, v.43;No.258(9): 102-105.
[4] 张强,王海舰,井旺,等. 基于模糊神经网络信息融合的采煤机煤岩识别系统[J]. 中国机械工程, 2016, 27(2): 201-208.
Zhang Qiang, Wang Haijian, Jing Wang, et al. Shearer’s Coal-rock Recognition System Based on Fuzzy Neural Network Information Fusion [J]. China Mechanical Engineering, 2016, v.27;No.434(2): 63-70.
[5] 刘俊利. 基于ANFIS的多信息融合煤岩识别方法研究[J]. 中国煤炭, 2014, 40(12): 56-59.
Liu Junli. Research on the coal-rock interface recognition based on the fusion of multi-information with ANFIS[J]. China Coal, 2014, v.40;No.461(12): 56-59.
[6] 李舜酩,郭海东,李殿荣. 振动信号处理方法综述[J]. 仪器仪表学报, 2013(8): 1907-1915.
Shun ming, Li Haidong, Li Dianrong. Review of vibration signal processing methods[J]. Chinese Journal of Scientific Instrument, 2013, (8): 1907-1915.
[7] 李秀坤,吴玉双. 多分量线性调频信号的Wigner-Ville分布交叉项去除[J]. 电子学报, 2017, 45(2): 315-320.
LI Xiukun, WU Yushuang. Cross-Term Removal of Wigner-Ville Distribution for Multi-component LFM Signals[J]. Acta Electronica Sinica, 2017, 45(2): 315-320.
[8] 王金贵,张苏. 基于频域约束独立成分分析的经验模态分解去噪方法[J]. 煤炭学报, 2017, 42(3): 621-629.
WANG Jin-gui, ZHANG Su. EMD denoising method based on frequency domain constrained independent component analysis[J]. Journal of China Coal Society, 2017, 42(3): 621-629.
[9] 郑红,周雷,杨浩. 基于小波包分析与多核学习的滚动轴承故障诊断[J]. 航空动力学报, 2015, 30(12): 3035-3042.
Zheng Hong, Zhou Lei, Yang Hao. Rolling bearing fault diagnosis based on wavelet packet analysis and multi kernel learning[J]. Journal of Aerospace Power, 2015, 30(12): 3035-3042.
[10] Liu JXLH. A Robust Approach of Watermarking in Contourlet Domain Based Onprobabilistic Neural Network[J]. Multimedia Tools and Applications, 2017, Vol.76(No.22): 24009-24026.
[11] 陈如清,李嘉春,尚涛,等. 改进烟花算法和概率神经网络智能诊断齿轮箱故障[J]. 农业工程学报, 2018, 34(17): 192-198.
Chen Ruqing, Li Jiachun, Shang Tao, et al. Intelligent fault diagnosis of gearbox based on improved fireworks algorithm and probabilistic neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(17): 192-198.
[12] 郭谋发,徐丽兰,缪希仁,等. 采用时频矩阵奇异值分解的配电开关振动信号特征量提取方法[J]. 中国电机工程学报, 2014, v.34;No.507(28): 4990-4997.
Guo Moufa, Xu Lilan, Miao Xiren, et al. A Vibration Signal Feature Extraction Method for Distribution Switches Based on Singular Value Decomposition of Time-frequency Matrix[J]. Proceedings of the CSEE, 2014, v.34;No.507(28): 4990-4997.
[13] 黄大荣,陈长沙,孙国玺,等. 复杂装备轴承多重故障的线性判别分析与反向传播神经网络协作诊断方法[J]. 兵工学报, 2017, 38(8): 1649-1657.
Huang Darong, CHEN Changsha, SUN Guoxi, et al. Linear Discriminant Analysis and Back Propagation Neural Network Cooperative Diagnosis Method for Multiple Faults of Complex Equipment Bearings[J]. Acta Armamentarii, 2017, 38(8): 1649-1657.
[14] 张国亮,杨春玲,王暕来. 基于优化概率神经网络和红外多光谱融合的大气层外空间弹道目标识别[J]. 电子与信息学报, 2014, 36(4): 896-902.
Zhang Guoliang, Yang Chunling, Wang Jianlai. Discrimination of Exo-atmospheric Targets Based on Optimization of Probabilistic Neural Network and IR Multispectral Fusion[J]. Journal of Electronics & Information Technology, 2014, 36(4): 896-902.
[15] 吴冲,刘佳明,郭志达. 基于改进粒子群算法的模糊聚类-概率神经网络模型的企业财务危机预警模型研究[J]. 运筹与管理, 2018, 27(2): 106-114, 132.
Wu Chong, Liu Jiaming, Gou Zhida. Use of Hybrid Fuzzy c-means and Probabilistic Neural Network Based on Improved Particle Swarm Optimization in the Prediction of Financial Distress[J]. Operations Research and Management Science, 2018, 27(2): 106-114, 132.
[16] 罗浩然,尹成,丁峰,等. 概率神经网络的平滑参数分析及在地震属性分析中的应用[J]. 石油物探, 2017, 56(4): 551-558.
Luo Haoran, Yin Cheng, Ding Feng. The smoothing parameter analysis of probabilistic neural network andits application in seismic attribute analysis [J]. Geophysical Prospecting for Petroleum, 2017, 56(4): 551-558.
[17] 丁青锋,尹晓宇. 差分进化算法综述[J]. 智能系统学报, 2017, 12(4): 5-16.
Ding Qingfeng, Yin Xiaoyu. Research survey of differential evolution algorithms[J]. CAAI Transactions on Intelligent Systems, 2017, 12(4): 5-16.
[18] 杨启文,蔡亮,薛云灿. 差分进化算法综述[J]. 模式识别与人工智能, 2008(4): 506-513.
Yang Qiwen, Cai Liang, Xue Yuncan. A Survey of Differential Evolution Algorithms [J]. Pattern Recognition and Artificial Intelligence, 2008, (4): 84-91.
[19] 周水生,周艳玲,姚丹,等. 基于QR分解的稀疏LSSVM算法[J]. 吉林大学学报(理学版), 2018, 56(2): 347-354.
Zhou Shuisheng;Zhou Yanling;Yao Dan, et al. Sparse LSSVM Algorithm Based on QR Factorization[J]. Journal of Jilin University(Science Edition), 2018, 56(2): 347-354.
[20] 刘江,李淼. 一种前向神经网络定值子集训练算法[J]. 信息系统工程, 2016, No.269(5): 12-14.
Li Jiang, Li Miao. A Forward Neural Network Fixed Value Subset Training Algorithm[J]. China CIO News, 2016, No.269(5): 12-14.
[21] 高大启. 有教师的线性基本函数前向三层神经网络结构研究[J]. 计算机学报, 1998(1): 80-86.
Gao Daqi. On structures Of supervised linear basis function feedforward three-layered neural networks[J]. Chinese Journal Of Computers, 1998, 21(1): 80-86
