Abstract
<title>Abstract</title> <p>This paper proposes a green building energy efficiency evaluation system suitable for different environments to meet the new requirements for buildings. The system will always be operational. Plan to build a modular computing platform to integrate dynamic, static, and hybrid models. Automatically preprocess data and adaptively orchestrate models in real-time to ensure accurate performance evaluation, thereby generating a dataset that includes twelve commercial buildings. These buildings have different uses and are suitable for different climates. The annual energy consumption estimation error rate of the hybrid model does not exceed 6%. The annual energy consumption estimation error rate for dynamic calculation is approximately 10%, while the annual energy consumption estimation error rate for static methods exceeds 14%. Even if the main input factors change by more than 10%, the corresponding changes in output factors are usually within 5%; the system is relatively stable. This system is an advanced precision tool used for energy comparison and evaluation in green building management. Relatively convenient and stable, it can be used for optimizing single buildings and large-scale investment portfolios.</p>