基于改进的动态因果图推理算法的复杂工业过程运行状态在线评价
Online assessment of complex industrial processes operating performance based on improved dynamic causality diagram

复杂工业生产中, 为获得优质产品和更高经济效益, 需保证生产过程运行于“正常”且“优”的状态. 针对定量信息与定性信息同时存在的生产过程, 提出一种基于改进的动态因果图(dynamic causality diagram, DCD)的过程运行状态评价方法. 该方法以因果图理论为基础, 扩充了定量分析和定性分析结合描述的动态因果图; 针对复杂工业生产过程中存在多个相互关联的事件联合作用导致一个事件发生的特点, 提出用联合因果强度来描述事件间关系的方法. 最后以金湿法冶金氰化浸出过程为应用背景, 分别利用传统DCD与改进DCD两种方法建立评价模型,对比分析评价结果，验证了文中提出方法的有效性和先进性.
For complex industry production, good operating performance is a prerequisite for high profits, low costs and so on. In order to grasp the process operating performance in real time, an online assessment method for process operating performance is necessary. Faced with hybrid of qualitative and quantitative analysis, dynamic causality diagram (DCD) is improved in this paper. To reduce the losses of quantitative messages, the certain information which is detected from the production field online will be fused with the uncertain information. Besides, the complex nonlinear relations among variables are common in chemical production. To address this issue, a new concept of Joint Event which describes one fluctuated event that must be impacted by multiple non-independent events is proposed in this paper. The modeling data are utilized to develop the assessment model via improved DCD and get accurate results during online assessment. Finally,the proposed method is applied to a hydrometallurgical leaching process operating performance assessment to illustrate its validity and effectiveness.