隐性扰动下智能车间资源重调度决策方法研究  

Decision method for intelligent workshop resource rescheduling under implicit disturbance

作　　者：苑明海[1,2] 黄涵钰 蔡仙仙 李子晨 裴凤雀 YUAN Minghai;HUANG Hanyu;CAI Xianxian;LI Zichen;PEI Fengque(College of Mechanical and Electrical Engineering,Hohai University,Changzhou 213022,CHN;Changzhou Key Laboratory of Intelligent Manufacturing Technology and Equipment,Changzhou 213022,CHN)

机构地区：[1]河海大学机电工程学院,江苏常州213022 [2]常州市智能制造技术与装备重点实验室,江苏常州213022

出　　处：《制造技术与机床》2023年第9期102-108,137,共8页Manufacturing Technology & Machine Tool

基　　金：江苏省自然科学基金面上项目(BK20201162);教育部人文社科规划基金项目(21YJA630111);常州市科技项目(CJ20220207);常州市科技计划项目(CM20223014)。

摘　　要：针对智能制造车间资源重调度快速响应的需求,以及隐性扰动难以测量捕捉的特点,提出了车间资源监测及重调度决策方法。首先利用支持向量机良好的连续监测性能,建立了资源异常状态监测模型;其次通过结合lasso回归算法和K近邻值分类算法提高SVM模型的预测值精准度,利用数据替代手段构建容错机制,保证系统异常时的短暂平稳运行;然后设计了车间资源重调度方式,通过历史案例数据训练分类器用于重调度方案抉择,指导智能制造车间在隐性扰动情况下的高效生产;最后,以实际车间隐性扰动为例,验证了所提的重调度决策方法的有效性。In order to meet the requirement of rapid response of resource rescheduling in intelligent manufacturing shop and the characteristics of invisible disturbance difficult to measure and capture,a decision-making method of resource monitoring and rescheduling in intelligent manufacturing shop was proposed.Firstly,a resource abnormal state monitoring model is established based on the good continuous monitoring performance of support vector machine.Secondly,the accuracy of SVM model was improved by combining Lasso regression algorithm and K-nearest neighbor value classification algorithm,and the fault-tolerant mechanism was constructed by data substitution method to ensure the transient smooth operation of the system in case of anomalies.Then,the workshop resource rescheduling method is designed,and the classifier is trained for rescheduling scheme selection by historical case data to guide the efficient production of intelligent manufacturing workshop under the condition of invisible disturbance.Finally,the effectiveness of the proposed rescheduling decision method is verified by an example of actual workshop invisible disturbance.

关 键 词：智能制造 隐性扰动 车间调度 重调度决策 

分 类 号：TH165[机械工程—机械制造及自动化]