.png){kind=icon}
.png){kind=icon}
Return
Digital Twin Framework for Predictive Maintenance and Materials Selection Facing Climate Change
.png){kind=icon}
PRE
AI
DOI：10.1149/2.F09261IF.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The resilience and longevity of metallic infrastructures are increasingly challenged by the growing impacts of climate change, which manifest through complex and evolving environmental stressors. Traditional approaches to corrosion management, which have historically relied on empirical models, periodic inspections, and the extrapolation of historical climate data, are proving insufficient in the context of climate change, including rapidly changing weather conditions. This paper articulates a scientific vision for the future of predictive maintenance and materials selection for metallic structures exposed to dynamic climate regimes. Central to this vision is the development of an integrated digital twin framework that couples high-resolution regional-to-local climate modeling with advanced corrosion prediction methodologies, encompassing both physics-based and data-driven approaches. The digital twin is dynamically updated through an active feedback loop incorporating real-time data from environmental and corrosion sensors, as well as periodic inspection records. In addition, the digital twin framework will be used to understand the consequences of future climate conditions. Designed to operate at spatial resolutions down to a few kilometers, the system enables site-specific risk assessments and supports informed decision-making for both maintenance scheduling and the selection of materials for new infrastructure. The framework further integrates scenario analysis, uncertainty quantification, and risk-based decision-making, thereby supporting the transition from reactive to predictive and adaptive asset management strategies.
Keywords:
corrosion
digital twin
predictive maintenance
materials science
climate change
Journal
IF:
1.4
Papers:
57
Citations:
1.0K