Advertisement

Heat and Mass Transfer

, Volume 43, Issue 8, pp 749–758 | Cite as

Experimental study and artificial neural network modeling of unsteady laminar forced convection in a rectangular duct

  • Nedim SözbirEmail author
  • İsmail Ekmekçi
Original

Abstract

In this study, an artificial neural network (ANNs) for the prediction of unsteady heat transfer in a rectangular duct was studied. The ANNs has been applied for the unsteady heat transfer in a rectangular duct. An experimental study has been carried out to investigate the axial variation of inlet temperature and the impact of inlet frequency on decay indices in the thermal entrance region of a parallel plate channel. The investigation was conducted with laminar forced flows for Reynolds numbers ranging from 1,120 to 2,200 while the inlet heat input frequency varied from 0.02 to 0.24 Hz. The results revealed that the ANNs can be used for modeling unsteady heat transfer in the duct. The accuracy between experimental and ANNs approach results was achieved with a mean absolute relative error less than 39%.

Keywords

Reynolds Number Hide Layer Test Section Mass Flow Rate Inlet Temperature 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

List of symbols

a

the width of the rectangular channel, m

b

half height of the rectangular channel, m

d

diameter of the orifice plate, m

De

equivalent diameter of rectangular test section or hydraulic diameter, m

Dp

diameter of PVC pipe, m

f

logistic sigmoid activation function

h

hidden layer

O

output

OL

output layer

m

mass flow rate, kg/s

p

design parameter (consequent parameter)

Pr

Prandtl number

q

design parameter (consequent parameter)

r

design parameter (consequent parameter)

t

time

T

temperature,°C

V

output voltage of thermocouple, mV

w

wiring strength of a rule

W

weights

x

axial distance, m

X

input

Y

target activation of the output layer

Z

dimensionless distance, Z=x/D e

Greek symbols

α

learning rate

β

inlet frequency, Hz

δ

error for output neuron

φ

orifice plate diameter ratio, d/D p

ρ

fluid density, kg/m3

μ

dynamic viscosity, Pa s

ΔP

pressure drop across the orifice plate, Pa

ΔTi

temperature amplitude at the center of the inlet,°C

θ

threshold between the input and hidden layers

η

momentum factor

Subscripts

I

input

max

maximum

min

minimum

o

output

p

value related to the redevelopment section

t

value related to the test section

amp

amplitude

References

  1. 1.
    Kakaç S, Li W, Cotta RM (1990) Unsteady laminar forced convection with periodic variation of inlet temperature. Trans ASME J Heat Transfer 112:913–920CrossRefGoogle Scholar
  2. 2.
    Li W, Kakaç S (1991) Unsteady thermal entrance heat transfer in laminar flow with a periodic variation of inlet temperature.Int J Heat Mass Transfer 34:2581–2592zbMATHCrossRefGoogle Scholar
  3. 3.
    Sozbir N (1995) Experimental investigation of unsteady forced convection in a rectangular channel with or without arrays of block-like electronic component, Report, University of Miami, Coral Gables, FL, USAGoogle Scholar
  4. 4.
    Sozbir N, Brown DM, Santos CAC, Kakac S, Guven HR (1995) Experimental Investigation of unsteady forced convection in a channel with and without arrays of rectangular protruding surfaces. J Therm Sci Technol 17(4): TIBTD Printed in Turkey, ISSN 1300–3615Google Scholar
  5. 5.
    Ekmekci I et al (2001) Investigation of forced convection in a duct with artificial neural network. In: 3rd International symposium on intelligent manufacturing systems, Sakarya, Turkey, pp 30–31Google Scholar
  6. 6.
    Hasiloglu A et al (2004) Adaptive neuro-fuzzy modeling of transient heat transfer in circular duct airflow. Int J Therm Sci 43:1075–1090CrossRefGoogle Scholar
  7. 7.
    Kalogirolu SA (2000) Applications of artificial neural networks for energy systems. Apply Energy 67:17–35CrossRefGoogle Scholar
  8. 8.
    Xu K et al (1998) Integration of neural networks and expert systems for microscopic wear particle analysis. Knowl-Based Syst 11:213–227CrossRefGoogle Scholar
  9. 9.
    ASME Standards (1984) Measurement of fluid flow in pipe using orifice, nozzle and venture, MCF-3M-1984Google Scholar

Copyright information

© Springer-Verlag 2006

Authors and Affiliations

  1. 1.Department of Mechanical EngineeringSakarya UniversitySakaryaTurkey

Personalised recommendations