| Acrylonitrile unit is an important chemical production for chemical enterprises. With the rapid development of the new chemical process, device design, new catalyst and materials, the modern acrylonitrile process and units are more and more large-scale and complicated. These developments improve the profit and competition competency of the enterprises; nevertheless, they need more safety managements and accident emergency controls. How to prevent from accident and diminish potential hazard to lower the accident probability receives much concern up to now. This thesis expatiated on a safety assessment to an in-service acrylonitrile unit of a chemical enterprise, which identified the hazard characteristics of the materials and process by risk gradation and ICI's mond methods, pointed the major hazard sources, and gave analogs to some major accident consequences. An emergency control study is conducted to the key point and potential hazard of the unit. Meanwhile, combined to the real-time detecting and monitoring system of the unit, an emergency control system is developed by picking up the data from the distributed control system (DCS) with OLE for process control (OPC) Client and flammable and toxic gas alarm server with adding hardwares. By means of least square support vector machine (LS-SVM) calculation, the software can predict the process data and compare with the threshold parameters, which function is emergency pre-warning. The emergency control system in this thesis focuses on the remote monitoring, real-time detecting, emergency pre-warning and accident consequence analysis. The system provides a technique support to the unit, which helps to control the risks, diminish accidents and damages rapidly and effectively in abnormal situation. The research gives a powerful reference to the onsite operation of the unit and production management of the related departments, which is of great significance to improve management efficiency and automation level for the unit.
|
PDF Full Text Request