Font Size: a A A

Study On The Test Method Of Crude Oil Moisture Content

Posted on:2018-07-28Degree:MasterType:Thesis
Country:ChinaCandidate:G D SunFull Text:PDF
GTID:2351330515954018Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Crude oil moisture content was an important data in crude oil production,and an important parameter index for oilfield development.In the process of production and storage of crude oil,the test of moisture content of crude oil was very important,which directly affected the exploitation,gathering,measurement,dehydration,sales and smelting of crude oil,which affected the quality and development cost of crude oil.Therefore,it was important to develop a crude oil moisture content tester with high test precision,wide test range,no radioactivity and low cost.In this paper,the research status and development trend of crude oil moisture content testing technology was introduced,and the commonly used water content test method of crude oil was studied and analyzed.The moisture content of the crude oil was measured by the density method,the advantages and disadvantages of density method and capacitance method was analyzed,the effects of temperature on the density method and the capacitance method were analyzed,designed of the density method and the capacitance of the test program,it is verified that the density error was small in the high water stage,and the measurement error was small in the low water stage.For density tested and capacitance tested,a multi-sensor fusion BP neural network model was proposed to predict the moisture content of crude oil,the neural network model of the output voltage of the pressure sensor,the output voltage value and the temperature value of the capacitance sensor as the input and the water content of the crude oil were constructed,a training sample was prepared under the premise of determining the temperature range and moisture content range,a representative sample was selected for the analysis of the results.The results show that the multi-sensor information fusion BP neural network model predicts the oil content of crude oil to make up for the shortcomings of density method and capacitance method,it had a high fitting accuracy and generalization ability,applied it to the crude oil moisture content test to more accurately obtain crude oil moisture content,in the oil moisture content of the test had a very good application value...
Keywords/Search Tags:Crude oil moisture content, Density method, Capacitance method, Multisensor information fusion, BP neural network
PDF Full Text Request
Related items