Font Size: a A A

Build The Process Neural Network Model For Pattern Recognition And Its Application In Discrimination Of Sedimentary Microfacies

Posted on:2018-12-24Degree:MasterType:Thesis
Country:ChinaCandidate:L Y LiuFull Text:PDF
GTID:2321330512997364Subject:Software engineering
Abstract/Summary:PDF Full Text Request
Due to the high cost and the small number of core wells,Daqing oilfield has many limitations by analyzing the core samples to identify sedimentary microfacies.In this paper,the data of a large number of logging curves of the core wells are combined with expert knowledge,research the pattern recognition method and application technology of the Process neural network,The results are applied to identify sedimentary microfacies of the non-core wells,which to achieve a certain block of oil reserves and mining value assessment.Combined with the actual analysis of the sedimentary microfacies found that-in addition to the digital data,the qualitative information such as expert experience may also have an effect on the discriminant results,and the microphase type itself is ambiguous and random,so if only use the quantitative identification model to identification may produce a lot of error.In this paper,fuzzy process neural network and hybrid computing process neural network based on cloud transform are used to identify the sedimentary microfacies.On the one hand,According to the rule data with expert experience,the fuzzy reasoning can be used to quantify the qualitative information of the logging phase to simplify the discriminant rule,extract the effective discriminant data and improve the accuracy of the sedimentary microfacies;On the other hand,for the data with fuzzy information,the cloud model can be used to transform the qualitative information in the phase data,which is included in the calculation process to ensure the integrity and objectivity of the original data;Considering that the data of the logging curve is the characteristic of the curve with depth,therefore,the process of neural network using the process of input advantage,with its time and space two-dimensional information processing capabilities and continuous optimization of the process neural network learning mechanism to improve the accuracy of sedimentation microfacies identification.Based on the data of sedimentary microfacies,the model and algorithm established by sedimentary microfacies identification are taken as the core,according to the development model of software engineering,the sedimentary microfacies model system was formed,and the experimental results were obtained.The research of the subject has a high practical value and its application prospect for the actual treatment of the sedimentary microfacies and the accurate expression of the microphase type.
Keywords/Search Tags:Pattern Recognition, Fuzzy Reasoning, Could Mdel, Process Neural Network, Mixed Computing
PDF Full Text Request
Related items