铝电解关键指标预测方法的研究与应用
|
摘要
为维持铝电解生产的持续性,保证电解槽的物耗稳定和能耗稳定,通过对铝厂数据的挖掘与建模,提出一整套维持电解槽稳定的策略方法,并用于指导实际生产。首先为数据去噪,针对铝厂数据分布特征未知的特点,提出一种无参自适应的模糊聚类方法,通过迭代自适应得到类簇个数和簇中心;根据聚类结果,将铝厂数据按实际意义标签化,提出一种基于距离的连续属性朴素贝叶斯算法,对分类器使用增量思想,使算法动态分类准确率得到提高;应用单槽测试集数据,通过累积法完成当天各指标等级趋势的预测,确定各指标下变量相对于前一天的变化量,完成预测。实验发现,预测模型可完成铝电解关键指标的预测;提出的聚类、分类算法在UCI数据及铝厂数据上表现良好。
This paper presents a strategy to maintain the stability of aluminum reduction cells through the data mining and modeling of the aluminum factory in order to maintain the continuity of aluminum electrolytic production and ensure the stability of energy consumption and the stability of energy consumption. Firstly, de-noising for data. In view of the unknown data distribution characteristics of aluminum plants, a fuzzy clustering method without parameter adaptation is proposed, and the number and cluster centers of clusters can be obtained through iterative adaptation. According to the clustering results, the data of aluminum factory is labelled according to the actual meaning. A distance based continuous attribute naive Bias algorithm is proposed. The incremental idea of classifiers is used to improve the accuracy of algorithm classification. By using single slot test set data, every index level can be predicted by accumulating method, compared with the previous day, the change of variables are confirmed under each index, and then complete the prediction. It is found that the prediction model can predict the key indicators of aluminum electrolysis, and the proposed clustering and classification algorithms are good in UCI data and aluminum factory data.
This paper presents a strategy to maintain the stability of aluminum reduction cells through the data mining and modeling of the aluminum factory in order to maintain the continuity of aluminum electrolytic production and ensure the stability of energy consumption and the stability of energy consumption. Firstly, de-noising for data. In view of the unknown data distribution characteristics of aluminum plants, a fuzzy clustering method without parameter adaptation is proposed, and the number and cluster centers of clusters can be obtained through iterative adaptation. According to the clustering results, the data of aluminum factory is labelled according to the actual meaning. A distance based continuous attribute naive Bias algorithm is proposed. The incremental idea of classifiers is used to improve the accuracy of algorithm classification. By using single slot test set data, every index level can be predicted by accumulating method, compared with the previous day, the change of variables are confirmed under each index, and then complete the prediction. It is found that the prediction model can predict the key indicators of aluminum electrolysis, and the proposed clustering and classification algorithms are good in UCI data and aluminum factory data.
引文
[1]胡丽青.模糊聚类在铝电解槽槽况分析中的研究与应用[D].北京:北方工业大学,2010.
[2]Zhang Hongliang,Yang Shuai,Li Jie,et al.Calculation method of voltage and heat balance of aluminum electrolytic tank:CN 103603010 A[P].2014.
[3]薛法远.基于槽况分类的铝电解电流效率预测研究[D].内蒙古包头:内蒙古科技大学,2016.
[4]农国武.基于BP神经网络的铝电解槽热平衡控制系统[J].轻金属,2006(8):47-50.
[5]黄迪,李太福,易军,等.基于优化相对主元分析的铝电解槽况诊断[J].计算机应用,2014,34(8):2429-2433.
[6]张旖芮,阳春华,朱红求.基于数据的铝电解槽况分类[J].计算机工程与应用,2015,51(11):233-237.
[7]聂轰.聚类与分类算法及其在铝电解数据分析中的应用研究[D].长沙:湖南大学,2008.
[8]林景栋,王丰,廖孝勇.基于小波神经网络的铝电解槽状态预测[J].控制工程,2012,19(2):290-293.
[9]杨炳儒,侯伟,徐章艳,等.一种基于关联分析的铝电解生产辅助控制方法[J].计算机应用研究,2009,26(2):631-633.
[10]王珊,刘胤,黄春杰,等.铝电解槽槽况综合分析方法研究[J].轻金属,2011(10):38-42.
[11]马忠林,赵旭东.基于经验的铝电解生产专家系统[J].轻金属,2006(2):33-35.
[12]付辉.模糊C-均值(FCM)聚类算法的改进[J].科学技术与工程,2007,7(13):3121-3123.
[13]王印松,商丹丹,宋凯兵,等.基于改进模糊聚类的控制系统故障检测[J].信息与控制,2017,46(1):41-45.
[14]肖满生,肖哲,文志诚,等.一种空间相关性与隶属度平滑的FCM改进算法[J].电子与信息学报,2017,39(5):1123-1129.
[15]Irpino A,Verde R,Carvalho F D A T D.Fuzzy clustering of distributional data with automatic weighting of variable components[J].Information Sciences,2017,406:248-268.
[16]Li R P,Mukaidono M.Maximum-entropy approach to fuzzy clustering[C]//IEEE International Conference on Fuzzy Systems,1995:2227-2232.
[17]Liang F,Xu Y,Li W,et al.Recognition algorithm based on improved FCM and rough sets for meibomian gland morphology[J].Applied Sciences,2017,7(2):192.
[18]张慧哲,王坚.基于初始聚类中心选取的改进FCM聚类算法[J].计算机科学,2009,36(6):206-209.
[19]许友权,吴陈,杨习贝.初始聚类中心优化的加权最大熵核FCM算法[J].计算机系统应用,2014,23(8):139-143.
[20]Karayiannis N B.MECA:maximum entropy clustering algorithm[C]//Proceedings of the Third IEEE Conference on Fuzzy Systems,2002:630-635.
[21]Miyamoto S,Umayahara K.Fuzzy clustering by quadratic regularization[C]//IEEE International Conference on Fuzzy Systems,2002:1394-1399.
[22]Wei C H,Fahn C S.The multisynapse neural network and its application to fuzzy clustering[M].[S.l.]:IEEE Press,2002.
[23]曾谁飞,张笑燕,杜晓峰,等.改进的朴素贝叶斯增量算法研究[J].通信学报,2016,37(10):81-91.
[24]黄琴.基于贝叶斯估计与分布的关联分析方法研究[D].广州:华南理工大学,2016.
[25]常青超.基于连续属性的贝叶斯分类方法应用研究[D].辽宁大连:大连海事大学,2016.
[26]John G H,Langley P.Estimating continuous distributions in bayesian classifiers[C]//Eleventh Conference on Uncertainty in Artificial Intelligence.[S.l.]:Morgan Kaufmann Publishers Inc,1995:338-345.
[27]Moriyama T.A new method of joint nonparametric estimation of probability density and its support[J].arXiv:1704.08015V3,2017.
[28]李丛,吴传生.一维连续随机变量概率密度估计[J].电子测试,2016(8):55-56.
[29]朱杰,陈黎飞.核密度估计的聚类算法[J].模式识别与人工智能,2017,30(5):439-447.
[30]熊开玲,彭俊杰,杨晓飞,等.基于核密度估计的K-means聚类优化[J].计算机技术与发展,2017,27(2):1-5.
[31]李俊林,符红光.改进的基于核密度估计的数据分类算法[J].控制与决策,2010,25(4):507-514.
[32]杜瑞杰,王双成,高瑞.基于高斯密度的一阶贝叶斯衍生分类器[J].计算机应用研究,2015(11):3242-3246.
[33]Tang B,He H.A local density-based approach for outlier detection[J].Neurocomputing,2017,241.
[2]Zhang Hongliang,Yang Shuai,Li Jie,et al.Calculation method of voltage and heat balance of aluminum electrolytic tank:CN 103603010 A[P].2014.
[3]薛法远.基于槽况分类的铝电解电流效率预测研究[D].内蒙古包头:内蒙古科技大学,2016.
[4]农国武.基于BP神经网络的铝电解槽热平衡控制系统[J].轻金属,2006(8):47-50.
[5]黄迪,李太福,易军,等.基于优化相对主元分析的铝电解槽况诊断[J].计算机应用,2014,34(8):2429-2433.
[6]张旖芮,阳春华,朱红求.基于数据的铝电解槽况分类[J].计算机工程与应用,2015,51(11):233-237.
[7]聂轰.聚类与分类算法及其在铝电解数据分析中的应用研究[D].长沙:湖南大学,2008.
[8]林景栋,王丰,廖孝勇.基于小波神经网络的铝电解槽状态预测[J].控制工程,2012,19(2):290-293.
[9]杨炳儒,侯伟,徐章艳,等.一种基于关联分析的铝电解生产辅助控制方法[J].计算机应用研究,2009,26(2):631-633.
[10]王珊,刘胤,黄春杰,等.铝电解槽槽况综合分析方法研究[J].轻金属,2011(10):38-42.
[11]马忠林,赵旭东.基于经验的铝电解生产专家系统[J].轻金属,2006(2):33-35.
[12]付辉.模糊C-均值(FCM)聚类算法的改进[J].科学技术与工程,2007,7(13):3121-3123.
[13]王印松,商丹丹,宋凯兵,等.基于改进模糊聚类的控制系统故障检测[J].信息与控制,2017,46(1):41-45.
[14]肖满生,肖哲,文志诚,等.一种空间相关性与隶属度平滑的FCM改进算法[J].电子与信息学报,2017,39(5):1123-1129.
[15]Irpino A,Verde R,Carvalho F D A T D.Fuzzy clustering of distributional data with automatic weighting of variable components[J].Information Sciences,2017,406:248-268.
[16]Li R P,Mukaidono M.Maximum-entropy approach to fuzzy clustering[C]//IEEE International Conference on Fuzzy Systems,1995:2227-2232.
[17]Liang F,Xu Y,Li W,et al.Recognition algorithm based on improved FCM and rough sets for meibomian gland morphology[J].Applied Sciences,2017,7(2):192.
[18]张慧哲,王坚.基于初始聚类中心选取的改进FCM聚类算法[J].计算机科学,2009,36(6):206-209.
[19]许友权,吴陈,杨习贝.初始聚类中心优化的加权最大熵核FCM算法[J].计算机系统应用,2014,23(8):139-143.
[20]Karayiannis N B.MECA:maximum entropy clustering algorithm[C]//Proceedings of the Third IEEE Conference on Fuzzy Systems,2002:630-635.
[21]Miyamoto S,Umayahara K.Fuzzy clustering by quadratic regularization[C]//IEEE International Conference on Fuzzy Systems,2002:1394-1399.
[22]Wei C H,Fahn C S.The multisynapse neural network and its application to fuzzy clustering[M].[S.l.]:IEEE Press,2002.
[23]曾谁飞,张笑燕,杜晓峰,等.改进的朴素贝叶斯增量算法研究[J].通信学报,2016,37(10):81-91.
[24]黄琴.基于贝叶斯估计与分布的关联分析方法研究[D].广州:华南理工大学,2016.
[25]常青超.基于连续属性的贝叶斯分类方法应用研究[D].辽宁大连:大连海事大学,2016.
[26]John G H,Langley P.Estimating continuous distributions in bayesian classifiers[C]//Eleventh Conference on Uncertainty in Artificial Intelligence.[S.l.]:Morgan Kaufmann Publishers Inc,1995:338-345.
[27]Moriyama T.A new method of joint nonparametric estimation of probability density and its support[J].arXiv:1704.08015V3,2017.
[28]李丛,吴传生.一维连续随机变量概率密度估计[J].电子测试,2016(8):55-56.
[29]朱杰,陈黎飞.核密度估计的聚类算法[J].模式识别与人工智能,2017,30(5):439-447.
[30]熊开玲,彭俊杰,杨晓飞,等.基于核密度估计的K-means聚类优化[J].计算机技术与发展,2017,27(2):1-5.
[31]李俊林,符红光.改进的基于核密度估计的数据分类算法[J].控制与决策,2010,25(4):507-514.
[32]杜瑞杰,王双成,高瑞.基于高斯密度的一阶贝叶斯衍生分类器[J].计算机应用研究,2015(11):3242-3246.
[33]Tang B,He H.A local density-based approach for outlier detection[J].Neurocomputing,2017,241.