Operating Performance Assessment Based on CVA-PRICA for Non-Gaussian Dynamic Industrial Processes

Accurate assessment of operating performance is critical to guarantee high economic benefit for industrial processes. In the traditional research works on assessment of operating performance, the static characteristics are fully investigated, but the dynamic characteristics are seldom explored. To fully extract the performance-related and dynamic features of non-Gaussian processes, this article presents a novel method (CVA-PRICA) which combines canonical variate analysis (CVA) and performance-relevant independent component analysis (PRICA) for operating performance assessment. The effectiveness of the proposed method is demonstrated through application to a gold hydrometallurgical process. The rates of accurately distinguishing the performance levels for CVA-PRICA is nearly 95% which illustrates the superiority of CVA-PRICA.