Load identification method of ball mill based on MEEMD- multi-scale fractal box dimension and ELM

doi:10.11949/j.issn.0438-1157.20180743

Abstract

Abstract:

In view of the problem that the load (filling rate and ball ratio) of a ball mill is difficult to be judged by experience during the grinding process, a method of mill load identification based on the multi-scale fractal box dimension of modified ensemble empirical mode decomposition (MEEMD) and extreme learning machine (ELM) is proposed. Firstly, the MEEMD algorithm is used to decompose the grind signals in different load states to get intrinsic mode components. Then, the correlation coefficient method is used to reconstruct the sensitive modal components to get the signal after noise reduction. By analyzing the multi-scale fractal box dimension of the reconstructed signal. The results show that there are obvious differences in the multi-scale fractal box dimensions of the under load, normal load and overloading state, and it can be well divided into different load states of the mill. The multi-scale fractal box dimension of regrinding signal is used as the input of extreme learning machine (ELM), and the load state of mill is output. The load identification model of mill is established. The effectiveness of the method is verified by grinding experiments. The recognition rate is as high as 94.8%, and the model can accurately identify the mill load status.

Key words: MEEMD, fractal box dimension, mill load, multi-scale, ELM

摘要：

针对球磨机在磨矿过程中负荷(充填率、料球比)靠经验难以准确判断的问题，提出基于改进的集总平均经验模态分解算法(modified ensemble empirical mode decomposition, MEEMD)-多尺度分形盒维数盒和极限学习机(extreme learning machine，ELM)的负荷识别方法。该方法首先利用MEEMD算法对不同负荷状态下的磨音信号进行分解得到本征模态分量，然后，采用相关系数法选取敏感模态分量进行重构得到降噪后信号；通过分析重构信号的多尺度分形盒维数，结果表明，欠负荷、正常负荷和过负荷状态下的多尺度分形盒维数存在明显的差异，能够很好地区分磨机的不同负荷状态。将重构磨音信号的多尺度分形盒维数作为极限学习机(ELM)的输入，磨机负荷状态为输出，建立磨机负荷识别模型；通过磨矿实验验证了该方法的有效性，整体识别率高达94.8%，模型能够准确识别磨机负荷状态。

关键词: MEEMD, 分形盒维数, 磨机负荷, 多尺度, ELM

CLC Number:

TP 29

Gaipin CAI, Lu ZONG, Xin LIU, Xiaoyan LUO. Load identification method of ball mill based on MEEMD- multi-scale fractal box dimension and ELM[J]. CIESC Journal, 2019, 70(2): 764-771.

蔡改贫, 宗路, 刘鑫, 罗小燕. 基于MEEMD-多尺度分形盒维数和ELM的球磨机负荷识别方法[J]. 化工学报, 2019, 70(2): 764-771.

Figures/Tables 11

Fig.1 Raw signal

Fig.2 Correlation coefficient of modal components

Fig.3 Reconfiguration of ground sound signal

Table1 Signal to noise ratio of original signal and reconstructed signal

信号类型

原始信号SNR/dB

重构信号SNR/dB

工况一

工况二

工况三

8.53

9.64

6.81

20.92

21.75

23.26

Fig.4 Processing results of two algorithms

Table2 SNR of three algorithms after processing

算法类型

SNR/dB

EMD

EEMD

MEEMD

10.57

15.19

23.26

Fig.5 Fitting curves of cloud similarity values for all samples under normal load

Fig.6 RMSE of number and activation function of different hidden neurons

Fig.7 Training recognition rate

Fig.8 Overall recognition rate of the number and activation function of different hidden neurons

Table 3 Recognition accuracy of different load states

负荷状态

识别准确率/%

欠负荷

正常负荷

过负荷

93.4

92.7

98.3

References 31

1	杜永贵, 李思思, 阎高伟, 等. 基于流形正则化域适应湿式球磨机负荷参数软测量[J]. 化工学报, 2018, 69(3): 1244-1251.
	DuY G, LiS S, YanG W, et al. Soft sensor of wet ball mill load parameter based on domain adaptation with manifold regularization[J]. CIESC Jorunal, 2018, 69(3): 1244-1251.
2	李景杰. 我国选矿机械设备大型化研究现状及发展动向[J]. 中国机械, 2014, 50(10): 52-57.
	LiJ J. Research status and development trend of large scale equipment for mineral processing in China [J]. Machine China, 2014, 50(10): 52-57.
3	ShaY, ChangT, ChangJ. Measure methods of ball mill's load[J]. Modern Electric Power, 2006, 24(35): 624-632.
4	ZhouP, ChaiT Y, WangH. Intelligent optimal-setting control for grinding circuits of mineral processing process[J]. IEEE Transactions on Automation Science & Engineering, 2009, 6(4): 730-743.
5	汤健. 基于频谱数据驱动的旋转机械设备负荷软测量[M]. 北京: 国防工业出版社, 2015: 23-50.
	TangJ. Load Measurement of Rotating Machinery Based on Spectrum Data Driving[M]. Beijing: National Defense Industry Press, 2015: 23-50.
6	汤健. 基于机械信号的磨矿过程磨机负荷检测综述[C]//第28届中国过程控制会议, 2017.
	TangJ. Summary of mill load detection based on mechanical signal for grinding process[C]// The twenty-eighth China Process Control Conference, 2017.
7	BhaumikA, BanerjeeS, SilJ. Designing of intelligent expert control system using petri net for grinding mill operation[J] .Wseas Transactions on Applicatons , 2005, 4(2): 147-150.
8	刘晓璞. 面向磨机负荷识别的新型磨音检测装置研究[D]. 济南: 济南大学, 2012.
	LiuX P. Research of the mill noise detection device for identification of the mill load[D]. Jinan: University of Jinan, 2012.
9	TangJ, YuW, ChaiT Y, et al. Selective ensemble modeling load parameters of ball mill based on multi-scale frequency spectral features and sphere criterion[J].Mechanical Systems and Signal Processing , 2016, 69(9): 485-504.
10	王飞. 基于频谱分析的磨机负荷检测方法研究[D]. 重庆: 重庆邮电大学, 2016.
	WangF. Research on the detection method of mill load based on spectrum analysis[D]. Chongqing: Chongqing University of Posts and Telecommunications, 2016.
11	惠瑜. 基于磨音的球磨机负荷软测量方法研究[D]. 西安: 西安理工大学, 2014.
	HuiY. Research on soft-sensing method of ball mill load by grinding sound[D]. Xi’an: Xi’an University of Technology, 2014.
12	汤健, 赵立杰, 岳恒, 等. 基于多源数据特征融合的球磨机负荷软测量[J]. 浙江大学学报(工学版), 2010, 44(7): 1406-1413.
	TangJ, ZhaoL J, YueH, et al. Soft sensor for ball mill load based on multi-source data feature fusion[J]. Journal of Zhejiang University(Engineering Science) , 2010, 44(7): 1406-1413.
13	ZhaoL J, YuanD C, TangJ. Ball mill load state recognition based on kernel PCA and probabilistic PLS-ELM[J]. Applied Mechanics and Materials, 2013, 26(1): 398-407.
14	汤健, 柴天佑, 丛秋梅, 等. 基于EMD和选择性集成学习算法的磨机负荷参数软测量[J]. 自动化学报, 2014, 40(9): 1853-1866.
	TangJ, ChaiT Y, CongQ M, et al. Soft sensor approach for modeling mill load parameters based on EMD and selective ensemble learning algorithm[J]. Acta Automatica Sinica, 2014, 40(9): 1853-1866.
15	赵立杰. 基于EEMD和iPLS的球磨机负荷参数集成建模方法[C]//第25届中国过程控制会议论文集.中国自动化学会过程控制专业委员会, 2014.
	ZhaoL J. Ensemble modeling of ball mill load parameters based on EEMD and iPLS[C]//Proceedings of the Twenty-fifth China Conference on Process Control, Chinese Association of Automation, 2014.
16	XiaoH H, XueT T, LIUG X. The disturbance signal detection method of power quality based on MEEEMD[C]// Proceedings of 2015 International Conference on Advances in Mechanical Engineering and Industrial Informatics, 2015, 12(7): 83-91.
17	LuiY, LinJ L, ChenK. A stable algorithm of box fractal dimension and its application in pore structure[J]. Rare Material and Engineering, 2015, 44(4): 800-804.
18	ZhengZ, JiangW, WangZ. Gear fault diagnosis method based on local mean decomposition and generalized morphological fractal dimensions[J]. Mechanism and Machine Theory, 2015, 91(10): 151-167.
19	BegumS, BaruaS, FillaR.Classification of physiological signals for wheel loader operators using multi-scale entropy analysis and case-based reasoning[J]. Expert Systems with Application, 2014, 41(2): 295-305.
20	LiuZ, ChaiT Y, YuW, et al. Multi-frequency signal modeling using empirical mode decomposition and PCA with application to mill load estimation[J]. Neurocomputing, 2015, 69(23): 392-402.
21	罗小燕, 陈慧明, 卢小江. 基于网络搜索与交叉验证的SVM磨机负荷预测[J].中国测试, 2017, 43(1): 131-135.
	LuoX Y, ChenH M, LuX J. Load forecasting of SVM mills based on network search and cross validation [J].China Measurement & Test, 2017, 43(1): 131-135.
22	VapnikV N. The nature of statistical learning theory[J]. IEEE Transactions on Neural Networks, 1995, 10(5): 988-999.
23	AhilaR, SadasivamV, ManimalaK. An integrated PSO for parameter determination and feature selection of ELM and its application in classification of power system disturbance[J].Applied Soft Computing, 2015, 32(11): 23-37.
24	张杰, 王建民, 杨志刚, 等. 模糊神经网络在磨机负荷控制中的应用[J]. 仪表技术与传感器, 2014, 24(5): 66-68+79.
	ZhangJ, WangJ M, YangZ G, et al. Application of fuzzy neural network in mill load control[J]. Instrument Technique and Sensor, 2014, 24(5): 66-68+79.
25	KhanateM, AhmadiH, OmidM. Feature-level fusion based on wavelet transform and artificial neural network for fault diagnosis of planetary gearbox using acoustic and vibration signals[J]. Insight, 2013, 55(6): 323-329.
26	SaJ J M, BackesA R. ELM based signature for texture classification[J]. Pattern Recognition, 2015, 51(7): 395-401.
27	HerasD B, ArguelloF, QuesadaB P. Exploring ELM-based spatial-spectral classification of hyperspectral images[J]. International Journal of Remote Sensing, 2014, 35(2): 401-423.
28	HuangG B, ZhouH, DingX. Extreme learning machine for regression and multiclass classification[J]. IEEE Transaction on Systems, man, and Cybernetics, Part B: Cybernetics, 2012, 42(2): 513-529.
29	TangJ, WangD H, ChaiT Y. Predicting mill load using partial least squares and extreme learning machines[J]. Soft Computing, 2012, 16(9): 1585-1594.
30	ChaiT Y, ZhangL Y, SUC Y, et al. An intelligent mill load switching control of the pulverizing system for an alumina sintering process[J]. IEEE Transaction on Control Systems Technology, 2012, 20(3): 677-687.
31	柴天佑. 复杂工业过程运行优化与反馈控制[J]. 自动化学报, 2013, 39(11): 1744-1757.
	ChaiT Y. Operational optimization and feedback control for complex industrial processes[J]. Acta Automatica Sinica, 2013, 39(11): 1744-1757.

[1]	Yu QIAN, Yaoxi CHEN, Xiaofei SHI, Siyu YANG. Big data analysis of solar energy fluctuation characteristics and integration of wind-photovoltaic to hydrogen system [J]. CIESC Journal, 2022, 73(5): 2101-2110.
[2]	CHU Fei, PENG Chuang, JIA Runda, CHEN Tao, LU Ningyun. Online prediction method of batch process product quality based on multi-scale kernel JYMKPLS transfer model [J]. CIESC Journal, 2021, 72(4): 2178-2189.
[3]	Wei GE, Chengxiang LI, Feiguo CHEN. Pseudo-particle modeling of multi-scale reaction-transport coupling [J]. CIESC Journal, 2021, 72(12): 5928-5935.
[4]	Xiaoyan LUO, Congcong DAI, Tiedong CHENG, Gaipin CAI, Xin LIU, Jishun LIU. Load identification method of ball mill based on improved EWT multi-scale entropy and KELM [J]. CIESC Journal, 2020, 71(3): 1264-1277.
[5]	Zuohua LIU,Hongjun WEI,Xia XIONG,Changyuan TAO,Yundong WANG,Fangqin CHENG. Chaotic mixing performance enhanced by rigid-flexible impeller with long-short blades in stirred tank [J]. CIESC Journal, 2020, 71(11): 5080-5089.
[6]	Zuohua LIU,Hongjun WEI,Xia XIONG,Changyuan TAO,Yundong WANG,Fangqin CHENG. Experiment and numerical simulation of chaotic mixing performance enhanced by perturbed rigid-flexible impeller in stirred tank [J]. CIESC Journal, 2020, 71(10): 4621-4631.
[7]	Yanbin HOU,Xianwen GAO,Xiangyu LI. Prediction for dynamic liquid level of sucker rod pumping using generation of multi-scale state characteristics in oil field production [J]. CIESC Journal, 2019, 70(S2): 311-321.
[8]	Enwei ZHI, Fei YAN, Mifeng REN, Gaowei YAN. Soft sensor of wet ball mill load parameters based on transfer variational autoencoder - label mapping [J]. CIESC Journal, 2019, 70(S1): 150-157.
[9]	Xia XIONG, Zuohua LIU, Deyin GU, Facheng QIU, Liang WANG, Changyuan TAO, Yundong WANG. Chaotic mixing process of fly ash in acid leaching tank intensified by rigid-flexible impeller [J]. CIESC Journal, 2019, 70(5): 1693-1701.
[10]	YANG Jian, MENG Xianyang, GAO Kehui, WU Jiangtao. Helmholtz equation of state for n-pentene [J]. CIESC Journal, 2018, 69(4): 1315-1323.
[11]	DU Yonggui, LI Sisi, YAN Gaowei, CHENG Lan. Soft sensor of wet ball mill load parameter based on domain adaptation with manifold regularization [J]. CIESC Journal, 2018, 69(3): 1244-1251.
[12]	QIU Facheng, LIU Zuohua, LIU Renlong, QUAN Xuejun, CHEN Jiaxing, GU Deyin, LI Bing, WANG Yundong. Chaotic mixing performance in rigid-flexible impeller stirred tank with eccentric air jet [J]. CIESC Journal, 2018, 69(2): 618-624.
[13]	RUI Zebao, JI Hongbing. Multi-scale effect and catalyst design in catalytic combustion of organic waste gas [J]. CIESC Journal, 2018, 69(1): 317-326.
[14]	SONG Sufang, HAO Zhenhua, DONG Libo, LI Junguo, FANG Yitian. Establishment of bubble-based EMMS/PFB model and its application on dense section of pressurized fluidized-bed [J]. CIESC Journal, 2017, 68(8): 2998-3005.
[15]	LI Kun, HAN Ying, LI Shenming, WANG Tong. OS-ELM-based hybrid online modeling for motor load torque of beam pumping units [J]. CIESC Journal, 2017, 68(6): 2465-2472.