Using Deep Reinforcement Learning to Generate Digital Representation Models of Traditional Dwellings and Optimize Energy-Saving Effects

With the continuous increase in global energy demand and the increasingly serious environmental problems, building energy conservation has become one of the important ways to achieve sustainable development. Especially in the field of traditional dwellings, how to improve their energy efficiency and energy-saving effects while maintaining cultural heritage and architectural style is still an urgent problem to be solved. To this end, this paper introduces a method based on PPO (Proximal Policy Optimization) to generate a digital representation model of traditional dwellings and optimize its energy-saving effect. First, a digital representation model of traditional dwellings is constructed, and the appearance, structure and internal environment characteristics of the building are accurately restored through 3D modeling technology. Then, a deep reinforcement learning algorithm is used to design a reward mechanism based on building energy consumption. By training intelligent agents, the energy-saving design of the building is continuously optimized under different environmental conditions, including building layout, window configuration, wall materials, etc. Finally, the optimization scheme is evaluated in combination with a simulation system to simulate the energy efficiency performance of the building under different conditions, and the parameters and strategies of the model are further adjusted. Among them, under typical climatic conditions, the annual comprehensive energy consumption of the optimized building is reduced by 14.6%. At the same time, the PPO algorithm outperforms the traditional optimization method in terms of convergence efficiency and resource consumption. These results verify the effectiveness and feasibility of deep reinforcement learning in building energy-saving optimization, and provide an intelligent, data-driven optimization method for the energy-saving transformation of traditional houses.