Application Of Fuzzy Petri Nets In Polymer Flooding Comprehensive Adjustment System

As one of new emerging technologies in improving recovery ration of oil, polymer flooding is widespreadly used in domestic and foreign oil field development. The Daqing oil field is the earliest one that develops polymer flooding industrialization at home. As the increasingly expansion of polymer flooding scale, the request for the adjudgement of polymer flooding is higher and higher. At present, the adjustment scheme is determained mainy on the experts'experiences because the factors that affect measures of polyer flooding are so fuzzy and uncertain. The knowledge expression of traditional expert system, low effeciency was caused by restraint and mutual impact of production. In addition, this method of expression is lack in parallel inference support. Therefore, in computer technology developing today, it is an inevitable trend of polymer flooding development to establish a set of man-machine interactive intellectualizations system.Based on the theory of Fuzzy Petri Nets, the representation of the fuzzy production system of weighted knowledge is discussed. The relation and a transformation algorithm between the representation of fuzzy production system and Fuzzy Petri Nets are presented. This paper presents a Weighted Fuzzy Petri Nets Decision-Making Model which is based on weighted fuzzy production rule aiming at the fuzziness and uncertain knowledge of polymer flooding comprehensive adjustment system. Based on the Weighted Fuzzy Petri Nets Model, an algorithm of formalized decision reasoning process is proposed, in which all the constraints in the reasoning rules are considered. It demonstrates that the decision reasoning process is accomplished more simply and quickly as the algorithm is adopted and efficiency of the reasoning is also improved due to the parallel operation ability of Petri nets. By analyzing the dynamic running mechanism of Fuzzy Petri Nets, a maintenance algorithm is presented. The back propagation algorithm of neural networks is introduced into the model of FPN. The FPN has obtained the same learning ability from BP network in neural networks. The learning and training of FPN has vital significance to the establishment, renewal and maintenance of the rule system of polymer flooding. With the object-oriented programming technology and the visual programming technique, a proto type system is implemented by Visual C++6.0 under Win32 platform.