Modeling of heat exchanger by using bio-inspired algorithm
Modelling of heat exchanger helps to define the error that occurs during the operation. Hence by optimizing it using genetic algorithm and particle swarm optimization, the error that occurred could be minimized and compared between both algorithms.The primary objective of this study was to obtain st...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Trans Tech Publications Inc
2014
|
| Subjects: | |
| Online Access: | http://dx.doi.org/10.4028/www.scientific.net/AMM.660.831 http://dx.doi.org/10.4028/www.scientific.net/AMM.660.831 http://eprints.uthm.edu.my/6427/1/Modeling_of_Heat_Exchanger_by_using_Bio%2DInspired_Algorithm.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Modelling of heat exchanger helps to define the error that occurs during the operation.
Hence by optimizing it using genetic algorithm and particle swarm optimization, the error that
occurred could be minimized and compared between both algorithms.The primary objective of this
study was to obtain structural model using Autoregressive Moving Average Exogenous (ARMAX)
equation. In this study, data from heat exchanger experiment was used to determine the parameter
of ARMAX equation. Using genetic algorithm (GA) and particle swarm optimization (PSO),
ARMAX parameters are optimized. Hence, the transfer function represents the plant for modelling.
Validation test used were autocorrelation and cross-correlation to validate between normalised data
input and error. Based on the result obtained, for GA, the input parameters are -0.000214, -
0.000728, -0.0020, and -0.000804 while the output parameters are -1.0000, -0.1783, -0.1473 and
0.3248. For PSO, the input parameters are 0.0104, -0.0122, -0.0067 and 0.0118 while the output
parameters are -0.4274, -0.1256, -0.1865 and-0.2614. From validation test, GA produced smoother
and effective result compared to PSO with less noise exists. |
|---|