학술논문

Wax deposition rate model for crude oil pipeline based on neural network
Document Type
Conference
Source
2010 Sixth International Conference on Natural Computation Natural Computation (ICNC), 2010 Sixth International Conference on. 2:760-762 Aug, 2010
Subject
Components, Circuits, Devices and Systems
Computing and Processing
Signal Processing and Analysis
Communication, Networking and Broadcast Technologies
General Topics for Engineers
Petroleum
Pipelines
Artificial neural networks
Computational modeling
Training
Stress
Biological system modeling
wax deposition rate
artificial neural network
crude oil pipeline
Language
ISSN
2157-9555
2157-9563
Abstract
In order to study the rule of wax deposition of Daqing oilfield crude oil in oil pipeline, an experimental unit is made by self. Based on experimental results, the relationship between the wax deposition rate and its influencing factors are established by using artificial neural network. The neural network is 4-7-1 BP structure, and the factors of influencing wax deposition rate include the shear stress, temperature gradient and molecular concentration gradient at pipeline wall, and dynamic viscosity of the crude oil. The predicted result shows that Prediction precision is high, error is less than 2%.