中文摘要：

为有效地从企业积累的制造过程质量数据中挖掘加工工序间的关联规则、实现制造过程的智能化监控,建立了基于质量数据融合和规则挖掘的离散制造过程监控模型。借鉴证据理论思想,提出一种基于模糊证据理论的多传感器质量数据融合算法,首先将所有测量值构成的集合视为辨识框架,应用模糊数学原理引入隶属度函数,设计了多传感器之间互支持度计算的新方法,并借鉴信任分配思想将测量值转换为相应的证据,最终通过基于冲突分配的证据组合规则得到融合结果,实现制造过程质量数据的精确采集。设计了基于质量规则挖掘的离散制造过程分析方法,针对传统关联规则挖掘结果繁杂且有不符合实际加工顺序规则生成的情况,借鉴有监督学习思想,以实际工序顺序为约束,挖掘导致产品质量问题的关键因素,结合复杂事件处理技术对制造过程进行实时监控。通过实验结果验证了该监控方法的有效性和实用性。

英文摘要：

To effectively extract the association rules from manufacturing process quality data accumulated by the enterprise and to realize the intelligent monitoring of manufacturing process,a discrete manufacturing process monitoring model based on quality data fusion and rule mining was developed.Drawing on evidence theory,a multi-sensor quality data fusion algorithm based on fuzzy evidence theory was proposed.The discernment frame was obtained from the collection of measured values,and a new method was provided to compute the mutual support among multisensors by using the membership function based on fuzzy mathematics principle.The concept of basic trust assignment was applied to generate the corresponding evidence,and finally the fusion result was obtained with the conflict assignment based combination rule,which realized the accurate collection for quality data of manufacturing process.A discrete manufacturing process analysis method based on quality rule mining was proposed.Aiming at the fact that the traditional association rule mining result was complex and did not accord with the actual processing sequence rule,the key factors leaded to the product quality problems were be mined in the actual process order constraint bylearning from the supervision of learning ideas.The manufacturing process was monitored in real time through the complex event processing technology.The effectiveness and practicability of monitoring method was indicated by the experimental results.