基于改进的高斯混合回归的球磨机料位软测量Soft measurement for ball mill fill level based on improved Gaussian mixture regression
杨飞,乔铁柱,庞宇松,阎高伟
摘要(Abstract)：
针对球磨机系统多模态复杂过程中的料位不确定性,球磨机振动信号存在非线性、噪声和外界干扰等问题,采用一种基于改进的高斯混合回归(GMR)的球磨机料位软测量方法,解决传统高斯混合模型初始化含有噪声和异常值的数据难以聚类的问题。首先,利用改进的K-medoids聚类算法与EM算法分别初始化和优化高斯混合模型(GMM)的最佳高斯分量个数、最优模型参数,然后采用GMR预测输出球磨机料位。最后实验验证了改进GMR模型得到的预测料位可以很好地跟踪真实料位,并且通过实验结果的对比分析,验证了改进模型的有效性和实用性以及较好的预测精度。
关键词(KeyWords)： 球磨机料位;多模态;振动信号;GMM;聚类;软测量;GMR
基金项目(Foundation): 国家自然科学基金项目(61450011);; 山西省自然科学基金项目(2015011052);; 山西省煤基重点科技攻关项目(MD 2014-07)~~
作者(Author): 杨飞,乔铁柱,庞宇松,阎高伟
DOI: 10.16652/j.issn.1004-373x.2018.05.035
参考文献(References)：
- [1]TANG Jian,ZHAO Lijie,YU Wen,et al.Soft sensor modeling of ball mill load via principal component analysis and support vector machines[J].Lecture notes in electrical engineering,2009(67):803-810.
- [2]汤健,郑秀萍,赵立杰,等.基于频域特征提取与信息融合的磨机负荷软测量[J].仪器仪表学报,2010,31(10):2161-2167.TANG Jian,ZHENG Xiuping,ZHAO Lijie,et al.Soft sensing of ball mill based on frequency domain feature extraction and information fusion[J].Chinese journal of scientific instrument,2010,31(10):2161-2167.
- [3]TANG Jian,WANG Dianhui,CHAI Tianyou.Predicting mill load using partial least squares and extreme learning machines[J].Soft computing,2012,16(9):1585-1594.
- [4]汤健,柴天佑,赵立杰,等.基于振动频谱的磨矿过程球磨机负荷参数集成建模方法[J].控制理论与应用,2012(2):184-186.TANG Jian,CHAI Tianyou,ZHAO Lijie,et al.Ensemble modeling for parameters of ball-mill load in griding process based on frequency spectrum of shell vibration[J].Control theory and applications,2012(2):184-186.
- [5]BISHOP C M.Pattern recognition and machine learning[M].New York:Springer,2007:432-443.
- [6]YU J.Multiway Gaussian mixture model based adaptive kernel partial least squares regression method for soft sensor estimation and reliable quality prediction of nonlinear multiphase batch processes[J].Industrial and engineering chemistry research,2012,51(40):13227-13237.
- [7]GRBI?R,SLI?KOVI?D,KADLEC P.Adaptive soft sensor for online prediction and process monitoring based on a mixture of Gaussian process models[J].Computers and chemical engineering,2013,58(22):84-97.
- [8]YUAN Xiaofeng,GE Zhiqiang,SONG Zhihuan.Soft sensor model development in multiphase/multimode processes based on Gaussian mixture regression[J].Chemometrics and intelligent laboratory systems,2014(138):99-103.
- [9]GHAHRAMANI Z,JORDAN M I.Supervised learning from incomplete data via an EM approach[J].Advances in neural information processing systems,1993(6):120-127.
- [10]KAUFMAN L,ROUSSEEUW P J.Finding groups in data:anintroduction to cluster analysis[M].New York:John Wiley&Sons,1990:108-123.
- [11]MCLACHLAN G,KRISHNAN T.The EM algorithm and extensions[M].New York:John Wiley&Sons,2007:105-132.
- [12]KULLBACK S.Information theory and statistics[M].US:Courier Dover Publications,2013:189-207.
- [13]周越,司刚全,曹晖,等.功率谱分析在筒式钢球磨煤机内存煤量测量中的应用研究[J].工业仪表与自动化装置,2006,35(6):21-24.ZHOU Yue,SI Gangquan,CAO Hui,et al.A study of the power spectrum analysis application to ball mill load measurement[J].Industrial instrumentation&automation,2006,35(6):21-24.
- [14]YU J,QIN S J.Multimode process monitoring with Bayesian inference-based finite Gaussian mixture models[J].Aiche journal,2008,54(7):1811-1829.
- [15]CHEN Xinquan,PENG Hong,HU Jingsong.K-medoids subatitution clustering method and a new clustering validity index method[C]//Proceedings of the 6th World Congress on Intelligent Control and Automation.[S.l.]:IEEE,2006:5896-5900.