Driven by global energy shortages and the “double carbon” goal, green buildings have become a central direction for the transformation of the construction industry. However, it is difficult to strike a balance between energy saving and human comfort place with problems such as slow response, insufficient accuracy and reliance on human experience. This article proposes a green building energy efficiency optimization and intelligent control system that integrates a number of artificial intelligence algorithms. It enables efficient energy consumption through a four-step approach: data acquisition, modeling, dynamic optimization and feedback adaptation. Firstly, it collects data on energy consumption, environmental parameters and user behaviour of buildings via multi-dimensional sensors. Second, LSTM models are built to predict energy consumption and enhanced learning optimization models, to accurately predict energy consumption trends and adapt control strategies. Finally, the dynamic unit corrects the operating parameters via a real-time feedback mechanism. In relation to total energy consumption in commercial buildings, the control group (commercial complexes) had the highest average daily energy consumption (218.4 kWh/day), energy consumption in the experimental group decreased to 174.5 kWh/day. Since these buildings include a number of commercial environments such as shopping malls and offices. The operating load of the equipment is high, and the traditional fixed PID control mode generates easily redundant energy consumption. The system proposed in this article achieves precise energy savings through dynamic adaptation.
Dong Wang (Thu,) studied this question.