@article{MAKHILLJEAS201914917750,
    title = {Multi-Objective Optimization of A HVAC System: Non-Dominated Sorting-Based
Differential Evolution Approach},
    journal = {Journal of Engineering and Applied Sciences},
    volume = {14},
    number = {9},
    pages = {3072-3082},
    year = {2019},
    issn = {1816-949x},
    doi = {jeasci.2019.3072.3082},
    url = {https://makhillpublications.co/view-article.php?issn=1816-949x&doi=jeasci.2019.3072.3082},
    author = {Y.N. and},
    keywords = {Energy,thermal comfort,optimization,multi-objective,promising,Non-dominating Sorting-based
Differential Evolution (NSDE)},
    abstract = {Energy efficiency of Heating Ventilating and Air Conditioning (HVAC) systems plays an important
role in reducing the world&#146;s energy needs. Optimization of HVAC system is one of the promising ways to
improve energy efficiency and to help in slowing down the depletion of our energy resources. This study
proposed an improved version of multi-objective optimization algorithm which integrates the simple yet
powerful differential evolution approach to the popular non-dominated sorting Genetic algorithm-II in solving
for optimization of a HVAC system. Energy consumption and thermal comfort are the two conflicting objectives
to be optimized with hourly cooling temperature set points of the thermal zone serve as the design variables
of the optimization. Optimization is performed through simulation of a case study using MATLAB coupled with
EnergyPlus Software. Cardinality, space and hyper volume metrics are used to measure and compare the quality
of the Pareto fronts obtained by the proposed algorithm, NSDE with the base algorithms, Differential Evolution
(DE) and Non-nominated Sorting Genetic Algorithm-II (NSGA-II). Decision making is also performed to evaluate
the energy performance of the proposed algorithm. Simulation results demonstrated that the NSDE produces
better quality Pareto optimal solutions compared to DE and NSGA-II as well as better energy saving capability
of the algorithm.}
    }