基于机器学习模型的选矿过程状态监测与故障诊断
|
摘要
提出并实现一种基于机器学习模型的选矿过程状态监测和故障诊断方法。基于通用的机器学习方法建立正常工况下的关键参数数据驱动模型;监测软件与DCS系统通信,实时计算目标变量的模型预测值并与实际测量值进行比较,误差超出设定阈值则进行报警标记;结合工艺专家的经验选择模型监测变量并与工况状态和工艺报警建立多方位联系,从而实现选矿过程状态监测和故障诊断。
In this paper,a kind of method for state monitoring and fault diagnosis of mineral processing based on machine learning model is put forward and realized. The data-driven models of key parameters of mineral processing under normal conditions are built based on general machine learning method. The system calculates the model predictive value of the target variable in real time through communication between the system and the mineral processing DCS. The real-time calculation results based on prediction model and the measured values of the target variable are compared. The alarm is marked when the error exceeds the set threshold. The monitoring variables are selected according to the experience of the process specialist. Multidimensional linkages between the alarm state and the working condition state or the process alarm are also established by expert experience. The method realizes the state monitoring and fault diagnosis of mineral processing.
In this paper,a kind of method for state monitoring and fault diagnosis of mineral processing based on machine learning model is put forward and realized. The data-driven models of key parameters of mineral processing under normal conditions are built based on general machine learning method. The system calculates the model predictive value of the target variable in real time through communication between the system and the mineral processing DCS. The real-time calculation results based on prediction model and the measured values of the target variable are compared. The alarm is marked when the error exceeds the set threshold. The monitoring variables are selected according to the experience of the process specialist. Multidimensional linkages between the alarm state and the working condition state or the process alarm are also established by expert experience. The method realizes the state monitoring and fault diagnosis of mineral processing.
引文
[1]王旭,徐宁,王庆凯,等.破碎流程故障诊断系统开发与应用[J].有色冶金设计与研究,2014,35(4):28.
[2]FRANK P M.Fault diagnosis in dynamic systems using analytical and knowledge-based redundancy:a survey and some new results[J].Automatica,1990,26(3):459.
[3]RAICH A C,CINAR A.Multivariate statistical methods for monitoring continuous processes:assessment of discrimination power of disturbance models and diagnosis of multiple disturbances[J].Chemometrics and Intelligent Laboratory Systems,1995,30(1):37.
[4]VENKATASUBRAMANIAN V,RENGASWAMY R,YINK,et al.A review of process fault detection and diagnosis[J].Computers&Chemical Engineering,2003,27(3):327.
[5]郭明.基于数据驱动的流程工业性能监控与诊断研究[D].杭州:浙江大学,2004.
[6]黄猛.基于大数据和机器学习模型的风力发电机组健康管理研究[J].机械制造,2017,55(8):37.
[7]牛晓晓.基于机器学习及智能算法的柴油机性能预测及优化研究[D].哈尔滨:哈尔滨工程大学,2017.
[8]沈林.基于深度学习的故障诊断方法研究[D].沈阳:沈阳理工大学,2018.
[9]张妮,车立志,吴小进.基于数据驱动的故障诊断技术研究现状及展望[J].计算机科学,2017,44(S1):37.
[10]VAPNIK V N.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000.
[11]叶涛.基于机器学习的软测量技术理论与应用[D].广州:华南理工大学,2007.
[2]FRANK P M.Fault diagnosis in dynamic systems using analytical and knowledge-based redundancy:a survey and some new results[J].Automatica,1990,26(3):459.
[3]RAICH A C,CINAR A.Multivariate statistical methods for monitoring continuous processes:assessment of discrimination power of disturbance models and diagnosis of multiple disturbances[J].Chemometrics and Intelligent Laboratory Systems,1995,30(1):37.
[4]VENKATASUBRAMANIAN V,RENGASWAMY R,YINK,et al.A review of process fault detection and diagnosis[J].Computers&Chemical Engineering,2003,27(3):327.
[5]郭明.基于数据驱动的流程工业性能监控与诊断研究[D].杭州:浙江大学,2004.
[6]黄猛.基于大数据和机器学习模型的风力发电机组健康管理研究[J].机械制造,2017,55(8):37.
[7]牛晓晓.基于机器学习及智能算法的柴油机性能预测及优化研究[D].哈尔滨:哈尔滨工程大学,2017.
[8]沈林.基于深度学习的故障诊断方法研究[D].沈阳:沈阳理工大学,2018.
[9]张妮,车立志,吴小进.基于数据驱动的故障诊断技术研究现状及展望[J].计算机科学,2017,44(S1):37.
[10]VAPNIK V N.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000.
[11]叶涛.基于机器学习的软测量技术理论与应用[D].广州:华南理工大学,2007.