Study On Synthetic Decision Support Systems For Petrochemical Enterprise Based On Intelligence Model Bases

Through the reform and opening-up of more than 20 years, especially today when our country has already become member of World Trade Organization, the management environments of enterprises of petroleum and petrochemical industry of our country have changed fundamentally too, the western trans-corporation has already tended towards the saturation because of its routine products market demand, expanding the space to the Asian countries actively, large quantities of international capital pour into peripheral the Far East, the Middle East and market of petroleum and petrochemical industry of country in Southeast Asia of our country, have strengthened the threat to petroleum petrochemical industry of our country. Meanwhile, after entering the WTO, the process of the internationalization of domestic market will be accelerated, the competition from trans- corporation and surrounding countries will be obviously aggravated, enterprises of petroleum and petrochemical industry should face more complicated management environment. So it is the inevitable choices of Chinese petroleum and petrochemical industry enterprises to solute to the challenge of new era change the strategic mode, promote the competitiveness.Petrochemical is the typical procedure enterprise, its production procedures are made up by related unit of a lot of complicated module inside which is successive and large scale. The automatic degree of petrochemical is relatively high, the technological progress is very fast, it is important ways to improve petrochemical enterprise's competitiveness though information-based traditional petrochemical of transformation, improving production efficiency, reducing the production cost, managing the informationization and making decisions intelligently.Analyse the design from the whole angle of petrochemical, the Synthetic Decision Support Systems for Petrochemical Enterprise (SDSS-PcE), which developed though Data Warehouse, On Line Analysis Processing and Data Mining combined, will become effective decision support tool for petrochemical in the economic globalization and information age.In addition, today when the crude oil price has already been integrated with the international, the frequent change of the international oil price causes very great uncertainty to the production costs of petrochemical of our country, the level of the crude oil price influences the downstream products of petroleum and cost and price that the petrochemical composed products directly, influencing the arrangement that petrochemical, risks of petrochemical industry enterprises have increased greatly too. Though petrochemical industry enterprises haye accumulated a large amount of international crude oil price data in process of production, but these data have not been fully utilized by the manager, studing how to excavate the international crude oil price change law from these data, being used for predicting the crude oil price, helping policymaker carry on decision, All those will become important meanings to petrochemicalThis thesis proceeds with production and management analysis of the petrochemical enterprise, has analysed in detail characteristics and contents of petrochemical enterprises production and management, offer the business support for the research of SDSS-PcE, Structure the Information Integrated Model for Petrolchemical Enterprise (IIM-PcE), has offered the information platform for SDSS-PcE.On the basis of the above two respects, this thesis structures the system architecture SDSS-PcE. SDSS-PcE system architecture is based on system architecture of DSS based on knowledge, including Language System, Problem Processing System, OLAP tool, Data Warehouse System,Knowledge Base System and Model Base System, the Model Base System is composed of Special Model Base System, Dynamic Model Base System and Intelligence Model Base System. The special model is the model designed to the concrete problem in advance; The dynamic model means that is used for producing the basic module of the new model and basic key element, it is form basic module and basic key element to divide up to go on some in common use model; The intelligence model means the algorithms or models of data mining. The Intelligence Model Base System is the key part of SDSS-PcE, make SDSS-PcE possess data analytical capacity of the depth, find knowledge ability and complicated computing capability through this system. The systematic design philosophy of Intelligence Model Base System is put forward for the first time by this thesis, has opened up new way and method for the research of SDSS.This thesis combining phase space reconstruction technology and recurrence plot technology, using artificial neural network stored in the Intelligence Model Base of SDSS-PcE, puts forward the artificial neural network prediction method based on phase space reconstruction technology. And structure the intelligence model of international crude oil price prediciton, has offered the new thinking and method for price prediciton of international crude oil, have offered the important basis for the fact of petrochemical enterprises production and management business decision too.