Estimation of flow rate through analysis of pipe vibration

Santhosh K. Venkata, Bhagya R. Navada

Research output: Contribution to journalArticle

Abstract

In this paper, implementation of soft sensing technique for measurement of fluid flow rate is reported. The objective of the paper is to design an estimator to physically measure the flow in pipe by analysing the vibration on the walls of the pipe. Commonly used head type flow meter causes obstruction to the flow and measurement would depend on the placement of these sensors. In the proposed technique vibration sensor is bonded on the pipe of liquid flow. It is observed that vibration in the pipe varies with the control action of stem. Single axis accelerometer is used to acquire vibration signal from pipe, signal is passed from the sensor to the system for processing. Basic techniques like filtering, amplification, and Fourier transform are used to process the signal. The obtained transform is trained using neural network algorithm to estimate the fluid flow rate. Artificial neural network is designed using back propagation with artificial bee colony algorithm. Designed estimator after being incorporated in practical setup is subjected to test and the result obtained shows successful estimation of flow rate with the root mean square percentage error of 0.667.

Original languageEnglish
Pages (from-to)294-300
Number of pages7
JournalActa Mechanica et Automatica
Volume12
Issue number4
DOIs
Publication statusPublished - 01-12-2018

Fingerprint

Pipe
Flow rate
Sensors
Neural networks
Backpropagation
Accelerometers
Mean square error
Amplification
Flow of fluids
Fourier transforms
Fluids
Liquids
Processing

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Mechanical Engineering

Cite this

@article{41f97593c9954f5ea400003287f6696e,
title = "Estimation of flow rate through analysis of pipe vibration",
abstract = "In this paper, implementation of soft sensing technique for measurement of fluid flow rate is reported. The objective of the paper is to design an estimator to physically measure the flow in pipe by analysing the vibration on the walls of the pipe. Commonly used head type flow meter causes obstruction to the flow and measurement would depend on the placement of these sensors. In the proposed technique vibration sensor is bonded on the pipe of liquid flow. It is observed that vibration in the pipe varies with the control action of stem. Single axis accelerometer is used to acquire vibration signal from pipe, signal is passed from the sensor to the system for processing. Basic techniques like filtering, amplification, and Fourier transform are used to process the signal. The obtained transform is trained using neural network algorithm to estimate the fluid flow rate. Artificial neural network is designed using back propagation with artificial bee colony algorithm. Designed estimator after being incorporated in practical setup is subjected to test and the result obtained shows successful estimation of flow rate with the root mean square percentage error of 0.667.",
author = "Venkata, {Santhosh K.} and Navada, {Bhagya R.}",
year = "2018",
month = "12",
day = "1",
doi = "10.2478/ama-2018-0045",
language = "English",
volume = "12",
pages = "294--300",
journal = "Acta Mechanica et Automatica",
issn = "1898-4088",
publisher = "Bialystok University of Technology",
number = "4",

}

Estimation of flow rate through analysis of pipe vibration. / Venkata, Santhosh K.; Navada, Bhagya R.

In: Acta Mechanica et Automatica, Vol. 12, No. 4, 01.12.2018, p. 294-300.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Estimation of flow rate through analysis of pipe vibration

AU - Venkata, Santhosh K.

AU - Navada, Bhagya R.

PY - 2018/12/1

Y1 - 2018/12/1

N2 - In this paper, implementation of soft sensing technique for measurement of fluid flow rate is reported. The objective of the paper is to design an estimator to physically measure the flow in pipe by analysing the vibration on the walls of the pipe. Commonly used head type flow meter causes obstruction to the flow and measurement would depend on the placement of these sensors. In the proposed technique vibration sensor is bonded on the pipe of liquid flow. It is observed that vibration in the pipe varies with the control action of stem. Single axis accelerometer is used to acquire vibration signal from pipe, signal is passed from the sensor to the system for processing. Basic techniques like filtering, amplification, and Fourier transform are used to process the signal. The obtained transform is trained using neural network algorithm to estimate the fluid flow rate. Artificial neural network is designed using back propagation with artificial bee colony algorithm. Designed estimator after being incorporated in practical setup is subjected to test and the result obtained shows successful estimation of flow rate with the root mean square percentage error of 0.667.

AB - In this paper, implementation of soft sensing technique for measurement of fluid flow rate is reported. The objective of the paper is to design an estimator to physically measure the flow in pipe by analysing the vibration on the walls of the pipe. Commonly used head type flow meter causes obstruction to the flow and measurement would depend on the placement of these sensors. In the proposed technique vibration sensor is bonded on the pipe of liquid flow. It is observed that vibration in the pipe varies with the control action of stem. Single axis accelerometer is used to acquire vibration signal from pipe, signal is passed from the sensor to the system for processing. Basic techniques like filtering, amplification, and Fourier transform are used to process the signal. The obtained transform is trained using neural network algorithm to estimate the fluid flow rate. Artificial neural network is designed using back propagation with artificial bee colony algorithm. Designed estimator after being incorporated in practical setup is subjected to test and the result obtained shows successful estimation of flow rate with the root mean square percentage error of 0.667.

UR - http://www.scopus.com/inward/record.url?scp=85060369303&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060369303&partnerID=8YFLogxK

U2 - 10.2478/ama-2018-0045

DO - 10.2478/ama-2018-0045

M3 - Article

VL - 12

SP - 294

EP - 300

JO - Acta Mechanica et Automatica

JF - Acta Mechanica et Automatica

SN - 1898-4088

IS - 4

ER -