Building Energy Systems Using Digital Twins and Genetic Algorithms

In order to solve the problems of energy consumption behavior and production process generating heat waste and carbon emissions, the author proposes to use digital twins and genetic algorithms to study building energy systems. The author employed Matlab/Simulink to develop an optimization framework for isolated multi-energy complementary building energy systems. The optimization objective was to minimize the annual cost of the system, and based on digital twins and genetic algorithms, the model was optimized and simulated for analysis. The experimental results show that compared to not considering flexible loads, when flexible electrical loads, flexible thermal loads, and flexible electrical/thermal loads participate in regulation, the annual cost of the system is reduced by 5.13%, 33.01%, and 35.4%, respectively. Incorporating flexible electrical loads into regulation shifts energy demand towards periods of high photovoltaic output, thereby reducing the required capacities of energy storage batteries and diesel generators. Compared to scenarios where only flexible thermal loads participate in regulation, simultaneous participation of both flexible electrical and thermal loads results in smoother indoor temperature fluctuations with reduced amplitude. When flexible thermal and electrical loads are simultaneously regulated, the best effect is achieved in reducing the annual value of system costs and annual carbon dioxide emissions.