The Research Of Chemical Process SDG Model Based On Complex Network

As the complexity of chemical process and the relationships between chemical production plants growing, the harsh conditions in operation flexibility, high temperature and pressure, vacuum cryogenic and corrosive poison of chemical industry is relatively serious. Fault diagnosis of chemical process is important for chemical industry safe operation. The application and research of chemical fault diagnosis theory based on SDG is in the ascendant. But, this method of fault diagnosis has some shortcomings, such as complex modeling, integrity is not high, easy to miss important variables, etc.At present, the theory of complex network is one of the effective methods to dispose complex and diversified issues. The research of complex network theory has been applied to fault diagnosis in the field of power grid and transportation, etc. The main goal of this project is to identify the key nodes of fault network and divide communities based on the theory of SDG chemical network diagnosis and complex network. So, in this paper, the problems of the SDG chemical network diagnosis will be solved and some researches have been carried out in SDG chemical network diagnosis.Firstly, the traditional flow chart of SDG modeling method was improved and a new method of modeling network adjacency matrix construction based on SDG model was put forward.Next, the model and method of key nodes recognition in chemical process diagnosis was put forward based on the combination of complex network and SDG The characteristics, centricity and communities of established chemical SDG network were calculated, judged and characterized. The important regions of chemical fault diagnosis SDG network were determined according to the community structure partitioning algorithm of complex network and the key nodes were recognised. The key nodes and path protection map was drew and applied to industrial examples to provide theoretical guidance for the safe operation of the network.At last, the propylene polymerization process, synthetic ammonia process and TE process as examples were analyzed and researched based on established model of complex network theory and SDG, the research focused on characteristics of the complex network, identification of key nodes and affirmance of key protected areas. The results show that the established model was feasible and had certain advantages on simplicity and accuracy than previous. This method proposed provides a new research idea and achievement for chemical production fault diagnosis.