A novel optimization algorithm based on epsilon constraint-RBF neural network for tuning PID controller in decoupled HVAC system

•Decoupling of a heating, ventilation, and air conditioning system is presented.•RBF models were identified by Epsilon constraint method for temperature and humidity.•Control settings derived from optimization of the decoupled model.•Epsilon constraint-RBF based on PID controller was implemented to...

Full description

Saved in:
Bibliographic Details
Published in:Applied thermal engineering 2016-04, Vol.99, p.613-624
Main Authors: Attaran, Seyed Mohammad, Yusof, Rubiyah, Selamat, Hazlina
Format: Article
Language:English
Subjects:
Algorithms
Computer simulation
Control systems
Controllers
Decoupled method
Epsilon constraint
Errors
HVAC
HVAC equipment
Neural network
Neural networks
Optimization
RBF
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Decoupling of a heating, ventilation, and air conditioning system is presented.•RBF models were identified by Epsilon constraint method for temperature and humidity.•Control settings derived from optimization of the decoupled model.•Epsilon constraint-RBF based on PID controller was implemented to keep thermal comfort and minimize energy.•Enhancements of controller parameters of the HVAC system are desired. The energy efficiency of a heating, ventilating and air conditioning (HVAC) system optimized using a radial basis function neural network (RBFNN) combined with the epsilon constraint (EC) method is reported. The new method adopts the advanced algorithm of RBFNN for the HVAC system to estimate the residual errors, increase the control signal and reduce the error results. The objective of this study is to develop and simulate the EC-RBFNN for a self tuning PID controller for a decoupled bilinear HVAC system to control the temperature and relative humidity (RH) produced by the system. A case study indicates that the EC-RBFNN algorithm has a much better accuracy than optimization PID itself and PID-RBFNN, respectively.
ISSN:1359-4311
DOI:10.1016/j.applthermaleng.2016.01.025