Treffer: Rule-based symbolic AI computing-driven digital twin model for ex-ante energy evaluation of task-ambience air conditioning systems in zero-energy buildings.

• Developed an ex-ante digital twin for TAACs energy evaluation in ZEBs. • Created rule-based symbolic AI to compute and optimize heat load in real time. • Built an indicator system linking thermal comfort with dynamic heat load data. • VR-integrated digital twin visualizes real-time energy and comfort feedback. • Model generates 48k standardized scenarios for design-phase decision support. The purpose of this study is to develop a rule-based symbolic AI computing-driven digital twin-based ex-ante energy evaluation model for Task-Ambience air conditioning systems (TAACs) in zero-energy buildings (ZEBs). This model introduces an interactive VR-based assessment framework, allowing designers to visually optimize energy efficiency and thermal comfort at the design phase, significantly improving decision-making efficiency compared to conventional static simulations. First, a comprehensive indicator system for ex-ante evaluation is established, incorporating thermal comfort and heat load as key parameters. The indicator system functions as the rule base of the symbolic AI computing framework, defining parameter ranges and logical conditions that govern heat load calculation and scenario generation, forming the core foundation of the model. A VR & rule-based symbolic AI computing simulation environment is then developed, enabling real-time environmental data acquisition, thermodynamic rule calculation, and visualization of energy consumption. "Rule-based symbolic AI computing" refers to embedding thermodynamic equations and predefined control rules into the digital twin for automated, real-time heat load calculation, providing immediate feedback for energy-efficient design adjustments. Validation shows an average energy-saving potential of 7.62% over conventional systems, confirming the model's effectiveness. The findings confirm that digital twin technology enhances pre-evaluation accuracy, reduces design-phase uncertainties, and improves decision-making efficiency for ZEBs. This study develops a globally applicable Visualized Ex-ante Evaluation Model (VEEM-ZEB), integrating thermal comfort and heat load indicators with rule-based symbolic AI computing in an interactive VR-based digital twin environment. Replacing traditional static simulations with a real-time, explainable, and climate-agnostic framework enables designers to balance energy efficiency and occupant comfort across diverse building types and regulatory contexts, while providing a scalable pathway for next-generation smart and sustainable building design worldwide. [ABSTRACT FROM AUTHOR]