Please wait a minute...
J4  2010, Vol. 44 Issue (7): 1406-1413    DOI: 10.3785/j.issn.1008-973X.2010.07.031
自动化技术     
基于多源数据特征融合的球磨机负荷软测量
汤健1, 赵立杰1,3, 岳恒2, 柴天佑1,2
1.东北大学 流程工业综合自动化教育部重点实验室,辽宁 沈阳 110189;2.东北大学 自动化研究中心,
辽宁 沈阳 110189;3. 沈阳化工大学 信息工程学院, 辽宁 沈阳 110142
Soft sensor for ball mill load based on multisource data
feature fusion
TANG Jian1, ZHAO Li-jie1,3, YUE Heng2, CHAI Tian-you1,2
1.Key Laboratory of Integrated Automation for Process Industry, Ministry of Education, Northeastern University,
Shenyang 110189, China; 2. Research Center of Automation, Northeastern University, Shenyang 110189, China;
3. College of Information Engineering, Shenyang University of Chemical Technology, Shenyang 110142, China
 全文: PDF  HTML
摘要:

针对磨矿过程球磨机负荷(ML)难以实时检测,生产中主要依靠人工经验判断负荷状态的难题,依据磨机筒体振动、振声、电流等信号与磨机负荷间存在相关性、信息互补与冗余的现象,提出基于多源数据特征融合的球磨机负荷软测量新方法.该方法由时域滤波、时频转换、特征提取、特征选择及软测量模型5部分组成.采用快速傅里叶变换(FFT)将滤波后的筒体振动及振声时域信号转换成频域信号,根据研磨机理将频域信号划分为低、中、高3个频段,采用核主元分析(KPCA)分别提取各个频段的非线性特征,选择振动、振声频域特征与电流时域特征的融合信号作为模型输入,建立基于最小二乘支持向量机(LSSVM)的磨机负荷软测量模型.实验结果表明,该方法与基于主元分析最小二乘支持向量机(PCALSSVM)方法和单传感器方法相比,磨机负荷参数预测精度较高.

Abstract:

The realtime measurement of ball mill load (ML) in grinding process is difficult to realize, and the states of ML are identified mainly by the experience of the operator. Aiming at the problems, a new softsensor approach of ML based on the multisource data feature fusion was proposed according to the relativity, the information complementation and redundancy among shell vibration, acoustic, electricity signal and ML. The approach consisted of five parts which were data filter, time/frequency transform, feature extraction, feature selection and soft sensor model. The shell vibration and acoustic signal in the time domain was transformed into the frequency domain using fast Fourier transform (FFT). The spectral signals were partitioned into three parts which were low, medium and high frequency bands according to the grinding mechanism. The kernel principal component analysis (KPCA) was used to extract the nonlinear feature of each part. The fused signals, which consisted of the frequency domain feature of vibration and acoustic signal, and the time domain feature of electricity signal, were selected as the input variables of the soft sensor model. The soft sensor model of ML was conducted based on the least square support vector machine (LSSVM). Experimental results show that the approach has better prediction accuracy for ML parameters than the PCALSSVM and the single sensor approaches.

出版日期: 2010-07-01
:  TP 29  
基金资助:

国家“863”高技术研究发展计划资助项目(2006AA060202).

通讯作者: 柴天佑,男,教授,院士.     E-mail: tychai@mail.neu.edu.cn
作者简介: 汤健(1974-),男,辽宁北票人,博士生,从事综合自动化系统及基于数据驱动技术的软测量建模研究. E-mail: tjian001@126.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  

引用本文:

汤健, 赵立杰, 岳恒, 柴天佑. 基于多源数据特征融合的球磨机负荷软测量[J]. J4, 2010, 44(7): 1406-1413.

SHANG Jian, DIAO Li-Jie, YUE Heng, CHAI Tian-You. Soft sensor for ball mill load based on multisource data
feature fusion. J4, 2010, 44(7): 1406-1413.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2010.07.031        http://www.zjujournals.com/eng/CN/Y2010/V44/I7/1406

[1] BHAUMIK A, SIL J, BANERJEE S. Designing of intelligent expert control system using petri net for grinding mill operation [J]. WSEAS Transactions on Applications, 2005, 4(2): 360365.
[2] NIEROP M A V, MOYS M H. Exploration of mill power modeled as function of load behaviour [J]. Minerals Engineering, 2001, 14 (10): 12671276.
[3] ZHOU P, CHAI T Y, WANG H. Intelligent optimalsetting control for grinding circuits of mineral processing [J]. IEEE Transactions on Automation Science and Engineering, 2009, 6(4): 730743.
[4] 周平,柴天佑.磨矿过程磨机负荷的智能监测与控制[J].控制理论与应用,2008,25(6): 10951099.
ZHOU Ping, CHAI Tianyou. Intelligent monitoring and control of mill load for grinding processes [J]. Control Theory and Applications, 2008, 25(6): 10951099.
[5] 白锐,柴天佑.基于数据融合与案例推理的球磨机负荷优化控制[J].化工学报,2009,60(7):17461751.
BAI Rui, CHAI Tianyou. Optimization control of ball mill load in blending process with data fusion and casebased reasoning [J]. Journal of Chemical Industry and Engineering, 2009, 60(7): 17461751.
[6] ZENG Y G, FORSSBERG E. Application of vibration signal measurement for monitoring grinding parameters [J]. Mechanical Systems and Signal Processing, 1994, 8(6): 703713.
[7] 李勇,邵诚.灰色软测量在介质填充率检测中的应用研究[J].中国矿业大学学报,2006,35(4):549555.
LI Yong, SHAO Cheng. Application research of grey soft sensor for charge ratio of media [J]. Journal of China University of Mining and Technology, 2006, 35(4): 549555.
[8] HUANG P, JIA M P, ZHONG B L. Investigation on measuring the fill level of an industrial ball mill based on the vibration characteristics of the mill shell [J]. Minerals Engineering, 2009, 22(14): 12001208.
[9] CAMPBELL J J, HOLEMES R J. The collection and analysis of single sensor surface vibration data to estimate operating conditions in pilotscale and productionscale AG/SAG mills [C]∥ Proceeding of 22nd International Mineral Processing Congress. CapeTown, South Africa: SAIMM, 2003: 280288.
[10] BEHERA B, MISHRA B K, MURTY C V R. Experimental analysis of charge dynamics in tumbling mills by vibration signature technique [J]. Minerals Engineering, 2007, 20 (1): 8491.
[11] SCHOLKOPF B, SMOLA A, MULLER K. Nonlinear component analysis as a kernel eigenvalue problem [J]. Neural Computation, 1998, 10(5): 12991319.
[12] MOGHADDAM B. Principal manifolds and probabilistic subspaces for visual recognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(6): 780788.
[13] 邹俊,傅新,黄硕,等.基于主成分分析的液压泥炮系统泄漏监测[J].浙江大学学报:工学版,2009,43(2): 197201.
ZOU Jun, FU Xin, HUANG Shuo, et al. Application of principal component analysis to monitoring of oil leakage in a hydraulic clay gun system [J]. Journal of Zhejiang University: Engineering Science, 2009, 43(2): 197201.
[14] 刘世成,王海清,李平.基于多向核主元分析的青霉素生产过程在线监测[J].浙江大学学报:工学版,2007,41(2): 202207.
LIU Shicheng, WANG Haiqing, LI Ping. Online monitoring of penicillin production process based on multiway kernel principal component analysis [J]. Journal of Zhejiang University: Engineering Science, 2007, 41(2): 202207.
[15] YU W, LI X O. Online fuzzy modeling via clustering and support vector machines [J]. Information Sciences, 2008, 178(22): 42644279.
[16] LIN S W, YING K C, CHEN S C. Particle swarm optimization for parameter determination and feature selection of support vector machines [J]. Expert System with Application, 2008, 35(4): 8171824.
[17] 林伟青,傅建中,许亚洲,等.基于最小二乘支持向量机的数控机床热误差预测[J].浙江大学学报:工学版,2008,42(6): 905908.
LIN Weiqing, FU Jianzhong, XU Yanzhou, et al. Thermal error prediction of numerical control machine tools based on least squares support vector machines [J]. Journal of Zhejiang University: Engineering Science, 2008, 42(6): 905908.
[18] 陈敏生,刘定平.基于核主元分析和支持向量机的电站锅炉飞灰含碳量软测量建模[J].华北电力大学学报,2006,33(1): 7275.
CHEN Minsheng, LIU Dingping. Softsensing modeling of the unburned carbon in fly ash based on KPCASVM for power station boilers [J]. Journal of North China Electric Power University, 2006, 33(1): 7275.
[19] 李哲,田学民.基于辅助变量KNN分析的软测量建模方法[J].化工学报,2008,59(4): 941946.
LI Zhe, TIAN Xuemin. Soft sensor modeling method based on secondary variables KNN analysis [J]. Journal of Chemical Industry and Engineering, 2008, 59(4):941946.
[20] 王泽红,陈炳辰.球磨机负荷检测的现状与发展趋势[J].中国粉体技术,2001,1(1): 1923.
WANG Zehong, CHEN Bingchen. Present state and development trend for ball mill load measurement [J]. China Powder Science and Technology, 2001, 1(1): 1923.

[1] 赵立杰, 柴天佑, 袁德成, 刁晓坤. 增强操作工况识别可靠性的概率PLS-ELM方法[J]. J4, 2013, 47(10): 1747-1752.
[2] 赵立杰, 汤健, 柴天佑. 基于选择性极限学习机集成的磨机负荷软测量[J]. J4, 2011, 45(12): 2088-2092.