Font Size: a A A

Analysis And Estimate Technology Of The Working Conditions Of ESP Well

Posted on:2017-02-09Degree:MasterType:Thesis
Country:ChinaCandidate:K YaoFull Text:PDF
GTID:2321330563451574Subject:Oil and gas engineering
Abstract/Summary:PDF Full Text Request
Electric submersible pump has the advantages of high efficiency,large displacement and high head,it plays an important role in non-flowing wells or some wells with high water cut,it becomes one of the necessary means to achieve the goal of high and stable production,meanwhile,it's the important part of energy consumption.Nowadays,most of oil fields is coming into the production tail,failures and problems are unavoidable,how to reduce the quantity and scale of damage seems especially important,this paper which about the analysis and estimate technology of the working conditions of the ESP well is based on this designation.First,this paper comprehends all kinds of failures of ESP well and there corresponding reasons.Then takes fuzzy evaluation method,the fuzzy neural network of current curve fault identification method based on PSO-BP algorithm,as well as the working-condition point and the hold-pressure-curve early warning analysis as the main topic.In order to establish a multi-angle working conditions analytical approach.Using fuzzy evaluation method to calculate the probability of failure according to the change trend of production data.The fuzzy neural network of current curve fault identification method based on PSO-BP algorithm can match the most probable fault type for the given current curve intelligently through the training of the sample.The working-condition point early-warning method can analysis the present working condition and take next steps according to the change trend of working-condition point.The hold-pressure-curve early warning analysis is mainly forecasts the well dead time of ESP wells by experiment,and then combines with other pump parameters to judge the fault and speculate the working condition.
Keywords/Search Tags:working conditions analysis, fuzzy evaluation, fault identification, current curve, fuzzy neural network, experiment of holding pressure
PDF Full Text Request
Related items