Font Size: a A A

Prediction Of Cementing Quality Based On Gray Fuzzy Neural Network And Safety Assessment

Posted on:2018-05-09Degree:MasterType:Thesis
Country:ChinaCandidate:C KongFull Text:PDF
GTID:2381330596969743Subject:Safety engineering
Abstract/Summary:PDF Full Text Request
Cementing operation is an extremely important part of drilling operation for oil and gas wells.The cementing quality will directly affect the normal exploitation of oil and gas wells and whether the late development can proceed smoothly,and it is closely related to the economic benefits of oil field exploration and development.So ensuring the high-quality cementing operation has been the focus of drilling construction in oil and gas wells.In view of the cementing quality prediction and the safety problems existing in the cementing operation,this paper mainly carried out the following work:Firstly,analyze the influence of different factors on the cementing quality from borehole conditions,displacement efficiency and cement slurry density through the literature review and field data.Through the grey relational grade analysis,we can conclude the main factors by sorting the factors that influence the cementing quality,which provides the theoretical basis for the establishment of cementing quality prediction method.Secondly,based on theories of gray system and fuzzy neural network,we can establish the grey method and fuzzy neural network method which can predict the cementing quality.The fuzzy neural network prediction method has higher accuracy than the gray method,but the training time is longer and the learning efficiency is lower.By combining the gray method with the fuzzy neural network method,a gray fuzzy neural network combining method for predicting the quality of cementing was established.Compared with the fuzzy neural network method,the proposed method can improve the accuracy of the prediction,the learning speed and generalization ability,and it can be used to guide the construction work.At the same time,using the software Visual Studio 2013,the database SQL Server 2008 and C# language can develop a set of software to achieve cementing quality prediction.Finally,through the risk identification analysis of dangerous and harmful factors hidden during cementing operation,we can build a set of scientific and reasonable evaluation index system and establish the membership function of the index.Based on the analytic hierarchy process,we can calculate the weight value of each evaluation index in the cementing operation system and establish the evaluation model to achieve the comprehensive safety evaluation of the cementing operation from the point view of the system,which can provide some reference for the safety management of oil field exploration and development.
Keywords/Search Tags:Gray system, Fuzzy neural network, Cementing quality, Prediction method, Safety evaluation
PDF Full Text Request
Related items