Information Fusion Method And Intelligent Reasoning Model Research Based On Cloud Transform

With the rapid development of computer technology, people have higher requirements onthe level of exploration and development of oil field, and they have more requirements on thecomprehensive utilization of various exploration-development data. It has been an importantresearch subject for oil workers to improve the level of understanding of petroleum depositand oil and gas production, through the comprehensive utilization of multidisciplinarydetection data of geology and logging involved in development geology and fusion researchwith qualitative knowledge of petroleum reservoir exploitation.Taking sedimentary microfiches identification and water-flooded identification as theapplication background, combined with the quantitative numerical information and qualitativeknowledge of oil fields, according to the information transformation mechanism of the cloudmodel, this paper has developed research on the fusion model and inference method of hybridinformation. Through the cloud model, it has interconverted the quantitative and qualitativeinformation involved in the identification problem, and then established quantitative andqualitative information fusion model; through reverse cloud transform, it has transformed theinput quantitative information to qualitative concept and established the correspondingrelation between input features and results of system. And then it has expressed the qualitativeconcept as the nerve cell, taken the learning nature of neural network to make inferentialanalysis, established the hybrid reasoning neural network based on reverse cloud transformand given the optimization algorithm based on Genetic algorithm; through the positive cloudtransform algorithm, it has transformed qualitative concept to quantitative information andthen taking the result as the input of neural network, it has established the hybrid calculationneural network based on positive cloud transform, and then it has taken the particle swarmoptimization of cloud variation to make optimization solution of network.Starting from the element data which can identify the sedimentary microfiches andwater-flooded layer, this subject has studied the analysis method, model and technology ofmulti-source data, and then applied the research results in the processing of actual data, whichhas yielded satisfactory research results and verified effectiveness of the model and method.