A Novel Lifecycle Operation Performance Evaluation Framework for Plant-Wide Industrial Processes

Abstract

In the modern complex industrial process, such as hot strip mill process (HSMP), the safety and optimality of the production process may deteriorate and operation performance will be degraded due to the wear of equipment, mode conversion and random disturbances. If the process is not adjusted and maintained, there may be more serious faults resulting in greater economic losses and potential safety hazards. Hence, it is essential to develop comprehensive operation performance evaluation. In this paper, a novel lifecycle operation performance evaluation framework based on multi-step total projection to latent structures (T-PLS) is proposed, which is used to deal with normal or faulty performance evaluate problems in the plant-wide HSMP. Firstly, HSMP can be divided into upstream, midstream and downstream by the actual process. Then, the T-PLS monitoring model is established gradually in each stream. Based on the pre-designed transmission rules of fault prior information, comprehensive statistical indexes are constructed to judge whether the process is in normal or faulty operation state. After that, the lifecycle performance evaluations are realized by different evaluation rules. Finally, the feasibility and effectiveness of the proposed method are illustrated through a real HSMP.