基于GA-SVR的挤压机能耗异常检测模型研究
|
摘要
挤压机作为铝型材生产的关键设备,其稳定性和可靠性是保障铝型材正常生产的前提。传统的设备异常检测方法属于侵入式检测,需要依靠外围仪器嵌入生产设备进行分析,检测成本高、自适应能力差。而能耗异常检测作为一种新颖的非侵入式检测手段,能够有效反映设备的运行工况,及时发现挤压过程的异常情况。为此将支持向量回归(SVR)与遗传算法(GA)相结合,提出一种基于GA-SVR的挤压机能耗异常检测模型。首先,根据挤压生产特点分析影响挤压能耗的关键因素,建立关键能耗因素为输入、电耗为输出的GA-SVR能耗预测模型。其次,考虑异常点的不确定性,基于GA-SVR模型构建单产能耗的置信区间并将其作为能耗异常区间。最后,以SY-1000Ton型挤压机为对象进行数值实验,验证所提模型的有效性。实验结果表明:置信度达到97%以上时所提模型检测能够准确地检测到挤压异常,这对保证挤压机稳定生产具有重要意义。
As the key equipment for the production of aluminum, the stability and reliability of aluminum extrusion machine is fundamental for normal production. Traditional equipment anomaly detection method belongs to intrusion detection, which requires other instrument to be embedded in production equipment for analysis, with the disadvantages of high-cost and low adaptive-capacity. As a new non-intrusion detection method, energy consumption anomaly detection can reflect operating conditions of equipment and find out anomaly situation timely. Therefore, support vector regression(SVR) and genetic algorithm(GA) were combined to provide an anomaly detection model. Through analyzing the key factors of extrusion energy consumption, a GA-SVR energy consumption prediction model was built, whose input was the key energy consumption factor and output was power consumption. Then, by considering the uncertainty of the anomaly point, a confidence interval of energy consumption was founded based on GA-SVR model and the confidence interval was seen as energy consumption anomaly interval. At last, an experiment was accomplished to verify the effectiveness of the model for SY-1000 Ton extruder. The results show that the model mentioned above can be used to find out anomaly accurately when confidence coefficient is more than 97%. So, it can be used to ensure the steady production of extruder.
As the key equipment for the production of aluminum, the stability and reliability of aluminum extrusion machine is fundamental for normal production. Traditional equipment anomaly detection method belongs to intrusion detection, which requires other instrument to be embedded in production equipment for analysis, with the disadvantages of high-cost and low adaptive-capacity. As a new non-intrusion detection method, energy consumption anomaly detection can reflect operating conditions of equipment and find out anomaly situation timely. Therefore, support vector regression(SVR) and genetic algorithm(GA) were combined to provide an anomaly detection model. Through analyzing the key factors of extrusion energy consumption, a GA-SVR energy consumption prediction model was built, whose input was the key energy consumption factor and output was power consumption. Then, by considering the uncertainty of the anomaly point, a confidence interval of energy consumption was founded based on GA-SVR model and the confidence interval was seen as energy consumption anomaly interval. At last, an experiment was accomplished to verify the effectiveness of the model for SY-1000 Ton extruder. The results show that the model mentioned above can be used to find out anomaly accurately when confidence coefficient is more than 97%. So, it can be used to ensure the steady production of extruder.
引文
[1] 刘静安,何梅琼,左其福,等.推广国产优质模具钢促进铝挤压工业与技术发展:SRM-1高强韧高温耐磨特种合金钢的特性及性价比分析[J].铝加工,2016(4):57-62.LIU J A,HE M Q,ZUO Q F,et al.Promotion of Domestic High-quality Steel,Promotion of Development of Aluminum Extrusion Industry and Technology:Characteristic and Cost Performance of SRM-1 Special Alloy Steel with High Strength and Toughness and High Temperature and Wear Resistance[J].Aluminium Fabrication,2016(4):57-62.
[2] 项胜前,罗铭强,陈文泗,等.铝合金一模多出(孔)挤压的特点及关键技术[J].铝加工,2015(5):9-14.XIANG S Q,LUO M Q,CHEN W S,et al.Characteristcs and Key Technical Analysis for Porous Mold Extrusion of Aluminum Alloy[J].Aluminium Fabrication,2015(5):9-14.
[3] 罗苏,吴锡坤.铝型材加工实用技术手册[M].长沙:中南大学出版社,2006.
[4] DIETMAIR A,VERL A.Energy Consumption Modeling and Optimization for Production Machines[C]//Proceedings of 2008 IEEE International Conference on Sustainable Energy Technologies,2008:574-579.
[5] 黄景涛,任志伟,罗威.电站锅炉监测数据的异常值检测算法研究[J].计算机与应用化学,2013,30(10):1153-1156.HUANG J T,REN Z W,LUO W.On Outliers Detecting Algorithm of Power Station Boiler Monitoring Data[J].Computers and Applied Chemistry,2013,30(10):1153-1156.
[6] 张宏利,刘吉成,李栋,等.基于免疫机理的设备异常度检测与故障诊断快速融合方法研究[J].机械设计与制造工程,2016,45(11):96-101.ZHANG H L,LIU J C,LI D,et al.Research on the Rapid Fusion Method for Equipment Abnormal Degree Detection and Fault Diagnosis Based on Immune Mechanism[J].Machine Design and Manufacturing Engineering,2016,45(11):96-101.
[7] 李楠.基于单类学习的异常检测及其重型装备状态识别应用[D].上海:上海交通大学,2014.
[8] 张纪.基于气味分析的设备异常检测方法研究[D].长沙:国防科学技术大学,2011.
[9] DRAKE R,YILDIRIM M B,TWOMEY J M,et al.Data Collection Framework on Energy Consumption in Manufacturing[C]//Proceedings of 2006 Institute of Industrial Engineering Research Conference,2006.
[10] MOUZONA G,MEHMET B,YILDIRIMA.A Framework to Minimize Total Energy Consumption and Total Tardinese on a Single Machine[J].International Journal of Sustainable Engineering,2007,1(2):105-116.
[11] 杨淑爱.制造过程中的低碳优化模型研究与应用[D].广州:华南理工大学,2012.
[12] MOUZONA G,MEHMET B,YILDIRIMA.Operational Methods for Minimization of Energy Consumption of Manufacturing Equipment[J].International Journal of Production Research,2007,45(3):4247-4271.
[13] EMEC S,KRüGER J,SELIGER G.Online Fault-monitoring in Machine Tools Based on Energy Consumption Analysis and Non-invasive Data Acquisition for Improved Resource-efficiency[J].Procedia Cirp,2016,40:236-243.
[14] 陀树青,梁鹏.一种挤压机能耗监测及预测系统研究[J].电子科技,2015(1):85-88.TUO S Q,LIANG P.Research on Energy Consumption Monitoring and Prediction System[J].Electronic Science and Technology,2015(1):85-88.
[15] VAPNIK V.The Nature of Statistical Learning Theory[M].New York:Springer,1995:24.
[16] 王雷,张瑞青,盛伟.基于支持向量机的回归预测和异常数据检测[J].中国电机工程学报,2009,29(8):92-96.WANG L,ZHANG R Q,SHENG W.Regression Forecast and Abnormal Data Detection Based on Support Vector Regression[J].Proceedings of The Chinese Society for Electrical Engineering,2009,29(8):92-96.
[17] CHUANG P J,HUANG S C,WU C F,et al.Chaos-based Support Vector Regressions for Exchange Rate Forecasting[J].Expert Systems with Applications,2010,37(12):8590-8598.
[18] RADOLPH G.Convergence Analysis of Canonical Genetic Algorithms[J].IEEE Transaction on Neural Network,1994,5(1):96-101.
[19] PARK C W,KWON K S,KIM W B,et al.Energy Consumption Reduction Technology in Manufacturing:A Selective Review of Policies,Standards,and Research[J].International Journal of Precision Engineering and Manufacturing,2009,10(5):151-173.
[2] 项胜前,罗铭强,陈文泗,等.铝合金一模多出(孔)挤压的特点及关键技术[J].铝加工,2015(5):9-14.XIANG S Q,LUO M Q,CHEN W S,et al.Characteristcs and Key Technical Analysis for Porous Mold Extrusion of Aluminum Alloy[J].Aluminium Fabrication,2015(5):9-14.
[3] 罗苏,吴锡坤.铝型材加工实用技术手册[M].长沙:中南大学出版社,2006.
[4] DIETMAIR A,VERL A.Energy Consumption Modeling and Optimization for Production Machines[C]//Proceedings of 2008 IEEE International Conference on Sustainable Energy Technologies,2008:574-579.
[5] 黄景涛,任志伟,罗威.电站锅炉监测数据的异常值检测算法研究[J].计算机与应用化学,2013,30(10):1153-1156.HUANG J T,REN Z W,LUO W.On Outliers Detecting Algorithm of Power Station Boiler Monitoring Data[J].Computers and Applied Chemistry,2013,30(10):1153-1156.
[6] 张宏利,刘吉成,李栋,等.基于免疫机理的设备异常度检测与故障诊断快速融合方法研究[J].机械设计与制造工程,2016,45(11):96-101.ZHANG H L,LIU J C,LI D,et al.Research on the Rapid Fusion Method for Equipment Abnormal Degree Detection and Fault Diagnosis Based on Immune Mechanism[J].Machine Design and Manufacturing Engineering,2016,45(11):96-101.
[7] 李楠.基于单类学习的异常检测及其重型装备状态识别应用[D].上海:上海交通大学,2014.
[8] 张纪.基于气味分析的设备异常检测方法研究[D].长沙:国防科学技术大学,2011.
[9] DRAKE R,YILDIRIM M B,TWOMEY J M,et al.Data Collection Framework on Energy Consumption in Manufacturing[C]//Proceedings of 2006 Institute of Industrial Engineering Research Conference,2006.
[10] MOUZONA G,MEHMET B,YILDIRIMA.A Framework to Minimize Total Energy Consumption and Total Tardinese on a Single Machine[J].International Journal of Sustainable Engineering,2007,1(2):105-116.
[11] 杨淑爱.制造过程中的低碳优化模型研究与应用[D].广州:华南理工大学,2012.
[12] MOUZONA G,MEHMET B,YILDIRIMA.Operational Methods for Minimization of Energy Consumption of Manufacturing Equipment[J].International Journal of Production Research,2007,45(3):4247-4271.
[13] EMEC S,KRüGER J,SELIGER G.Online Fault-monitoring in Machine Tools Based on Energy Consumption Analysis and Non-invasive Data Acquisition for Improved Resource-efficiency[J].Procedia Cirp,2016,40:236-243.
[14] 陀树青,梁鹏.一种挤压机能耗监测及预测系统研究[J].电子科技,2015(1):85-88.TUO S Q,LIANG P.Research on Energy Consumption Monitoring and Prediction System[J].Electronic Science and Technology,2015(1):85-88.
[15] VAPNIK V.The Nature of Statistical Learning Theory[M].New York:Springer,1995:24.
[16] 王雷,张瑞青,盛伟.基于支持向量机的回归预测和异常数据检测[J].中国电机工程学报,2009,29(8):92-96.WANG L,ZHANG R Q,SHENG W.Regression Forecast and Abnormal Data Detection Based on Support Vector Regression[J].Proceedings of The Chinese Society for Electrical Engineering,2009,29(8):92-96.
[17] CHUANG P J,HUANG S C,WU C F,et al.Chaos-based Support Vector Regressions for Exchange Rate Forecasting[J].Expert Systems with Applications,2010,37(12):8590-8598.
[18] RADOLPH G.Convergence Analysis of Canonical Genetic Algorithms[J].IEEE Transaction on Neural Network,1994,5(1):96-101.
[19] PARK C W,KWON K S,KIM W B,et al.Energy Consumption Reduction Technology in Manufacturing:A Selective Review of Policies,Standards,and Research[J].International Journal of Precision Engineering and Manufacturing,2009,10(5):151-173.