无规则数字孪生：基于标准化框架的声明式决策
原标题：Ruleless Digital Twins: Toward Declarative Decision-Making Through Standardized Frameworks and Technologies
AI 摘要
该论文提出「无规则数字孪生」（RDTs）方法，用纯声明式用户规格自动生成最优决策，替代传统基于规则的数字孪生决策模型。RDTs 结合语义知识库、逻辑推理、仿真模型和自主计算架构，且各组件均基于广泛使用的标准或技术。在虚拟办公室环境实验中，RDT 在维持目标室温和降低动态电价下能耗成本方面优于基于规则的 bang-bang 控制器，但组合决策树构建导致较长预测时域需要大量算力。
正文节选
Ruleless Digital Twins: Toward Declarative Decision-Making Through Standardized Frameworks and Technologies Abstract Digital twins (DTs) can be thought of as digital counterparts of physical objects, or more generally, twinning targets (TTs). To enact changes on and optimize for properties of their TTs, DTs use an element of decision-making. Traditionally, many domains, such as home automation, utilize rule-based decision-making models to imperatively define DT actions based on desired TT condit