Improving the information veracity of the complex of multiparametric control of the relaxometer based on a neural network

The article deals with studies of measurements of physico-chemical characteristics by a proton magnetic resonance relaxometer by methods of veracity control and operation control. The choice of the neural network structure is justified, the algorithm of training the neural network in the Statistica...

Full description

Saved in:
Bibliographic Details
Main Authors: Ovseenko, Galina A., Kashaev, Rustem S., Kozelkov, Oleg V., Filimonova, Tamara K., Evdokimova, Tatyana S., Mardanova, Aliya M.
Format: Conference Proceeding
Language:English
Subjects:
Artificial neural networks
control
Magnetic resonance
neural network
parameters
proton
Protons
relaxometry
Sociology
Software
Software algorithms
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The article deals with studies of measurements of physico-chemical characteristics by a proton magnetic resonance relaxometer by methods of veracity control and operation control. The choice of the neural network structure is justified, the algorithm of training the neural network in the Statistica 10 mathematical package is described according to the following parameters: spin-spin relaxation times, proton population and amplitude of spin-echo signals, which carry out a multiparametric analysis of fluid characteristics in digital intelligent deposits. This article solves the problem of increasing the information veracity of the complex of multiparametric control of the proton magnetic resonance relaxometer as part of the device-software package by developing and applying methods, algorithms and software based on them to evaluate the operating modes of the nodes of the complex of multiparametric control based on the use of artificial neural network technology.
ISSN:2831-7262
DOI:10.1109/REEPE57272.2023.10086740