Ontology Based Mathematical Knowledge Management In Petrochemical Industry

In recent decades, computer integrated manufacturing system (CIMS) brought great changes to all walks of life. During the application of CIMS to the process industry like petrochemical industry, computer integrated process system (CIPS) where process control and information system management are closely integrated has appeared. There is lots of mathematical knowledge in petrochemical enterprise. Mathematical knowledge was not managed effectively in CIPS, which results in waste of human resources and time during the process of enterprise software upgrade. In the petrochemical industry, mathematical knowledge management related research is still in its infancy. After analyzing mathematical knowledge and several existing management methods in petrochemical enterprises, we propose a new ontology based mathematical knowledge management method in which ontology-driven modeling theory is introduced.This thesis starts from the definition of mathematical knowledge and its management. Then, the current situation of mathematical knowledge management in petrochemical enterprises is described from the three levels of the CIPS. Later, the problems which occurred in the petrochemical enterprise mathematical knowledge management methods are analyzed. The ontology based mathematical knowledge management is introduced to solve the problems. OntoMODEL is one of the ontology based mathematical knowledge management methods which divides mathematical knowledge into declarative part and procedural part. OntoMODEL is a good solution to fulfill the articulation, organization, dissemination, access of the mathematical knowledge. However, OntoMODEL only can be applied to simple mathematical formulas based mathematical models, which is a vital limitation.To remedy the limitation of OntoMODEL, a new mathematical knowledge management is presented in this thesis. The method introduces the ontology driven modeling theory, uses instance of modeling ontology to represent mathematical knowledge, which separate the management and application of mathematical knowledge. In this thesis, the whole process is separated in three stages:model definition, model mapping and model usage. The proposed method uses ontology mapping to bridge the model definition and model usage, which can support complex non-mathematical formulas based mathematical model.DEVSMO and SMO are established in this thesis, which are the modeling ontology of the DEVS model and general optimization model. By using these two modeling ontology, a DEVS model and an optimization model are built as the modeling ontology instances and mapped to the executable model files. We also developed a mapping tool to translate the modeling ontology instance to executable code.Finally, this thesis implemented a typical petrochemical DEVS simulation by using the proposed ontology based mathematic knowledge management methods. The verification of the models and the simulation results proves the effectiveness of the proposed method.