A genetic algorithm for optimization of building envelope and HVAC system parameters

The aim of this paper is to describe the features of a Genetic Algorithm (GA) developed to solve simulation-based optimization problems for the optimal design of building parameters. This GA has been developed using guidelines from top researchs in the field of evolutionary computation. It is mostly based on NSGA-II and Omni-optimizer. It can be used for single and multi-objective optimization problems with and without constrains. Both discrete and continuous variables can be handled. Real-world optimization problems in the field of building performance simulation are carried out to verify the performance of the developed GA. Results for a single-objective optimization problem are presented, where the aim is minimization of life cycle cost of a detached house. Besides diverse sets of non-dominated solutions results for a multi-objective building design problem are also presented.