Operating performance of industrial process on safety and optimality may deteriorate with time due to process characteristic variation, and it is crucial to develop strategies for online operating performance assessment. Although there have been some studies and applications on process safety assessment, optimality assessment has not yet been paid sufficient attention. This paper proposes a probabilistic framework of online operating assessment for industrial processes. First, a Gaussian mixture model (GMM) is used to characterize multiple operating modes. Considering the distribution of process variables, safety and optimality indices (SI and OI) are defined and calculated by two successive nonlinear mappings. A hierarchical-level classification method is then presented to divide these indices into different performance levels, and margin analysis on each level is introduced. Finally, performance prediction and preliminary suggestions for improvement are provided. The proposed assessment strategy is then applied in two examples: Tennessee Eastman Process (TEP) and polypropylene (PP) production process, which indicate the efficiency of the proposed approach.
This paper proposes an approach of online safety assessment for industrial process operations. The core idea is to apply the Hopf bifurcation analysis of nonlinear systems in safety assessment. Herein, the Hopf bifurcation equations are used to characterize the critical stability boundary or manifold of the system, which is taken as the safety limit. Then, a safety index (SI) is constructed to denote the integrated exponential distances between each parameter and their bifurcation points. In the online implementation, the parameters are estimated by the extended Kalman filter (EKF), and the Hopf bifurcation points are generated by solving the nonlinear bifurcation equations. Afterward, the value of the SI can be online calculated. The introduced approach was then applied to a simulated gas phase polyethylene reactor process, in which the efficiency of the proposed method was verified in indicating the distance to the potential unsafe oscillation and in the early identification of potential threats.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.