Application of evolution strategies for HVAC optimization problem

Ideas of sustainability are not just for the design phase of buildings, but also extended to the stage of facilities management for daily operation. In the regime of strategic facilities management, to identify the economic energy conservation opportunities for optimal energy management becomes a primary objective for the centralized heating, ventilating and air-conditioning (HVAC) system of commercial building developments. Evolutionary algorithm (EA) has been broadly used in determining the required design and operating parameters of different engineering systems, which have the non-linear, discrete and highly constrained characteristics in the optimization scenarios. Generally, there are three main streams of EA for single objective optimization: Genetic algorithm (GA), evolutionary programming (EP) and evolution strategies (ES). In the past research works, GA has been widely adopted in the engineering optimization problems, but little studies about the effectiveness of EP and ES have been made. Therefore in this study, focus was directed on the performance appraisal of EP and ES, as well as the application potential for the engineering simulation problems. Conventional test functions were used to verify the effectiveness and efficiency of the two paradigms EP and ES, and the appropriate choice of different recombination operators. EP with arithmetic recombination was found to have the best optimization potential, and it was applied to an installed HVAC system to determine the optimal settings for better energy management.