When developing a digital twinDigital twin (DT) of a physical system, the model that the DT uses must be verified and validated. However, when developing verification and validationVerification and validation (V&V) strategies, the dynamic nature of the physical twin is not considered. Therefore, there are questions on whether the DT remains valid as the physical system undergoes changes during its lifetime, e.g., due to wear. In this paper, we introduce the use of Bayesian StatisticsBayesian statistics to dynamically update the DT as the data provided by the physical counterpart is received by the DT. This paper aims to advance the application of Digital TwinsDigital twin by improving V&V strategies.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Application of Bayesian Statistics to Refine Digital Twins

  • Samuel Cornejo,
  • Alejandro Salado

摘要

When developing a digital twinDigital twin (DT) of a physical system, the model that the DT uses must be verified and validated. However, when developing verification and validationVerification and validation (V&V) strategies, the dynamic nature of the physical twin is not considered. Therefore, there are questions on whether the DT remains valid as the physical system undergoes changes during its lifetime, e.g., due to wear. In this paper, we introduce the use of Bayesian StatisticsBayesian statistics to dynamically update the DT as the data provided by the physical counterpart is received by the DT. This paper aims to advance the application of Digital TwinsDigital twin by improving V&V strategies.