Predictive diagnosis of hidden risk for urban lifeline infrastructures driven by digital twin modeling of multisource observations: perspective

Researchers have represented a technical framework of the digital twin (DT) system based on multisource data fusion, which is for the predictive diagnosis of hidden risk for the entire life-cycle of urban lifeline infrastructures. Published in Smart Construction, this approach provides scientific support for the intelligent operation and maintenance of future cities.

The urban lifeline (urban transportation, water, electricity, oil, gas, central heating network) is similar to a city’s vascular network, and the infrastructures are the main carrier of various physical and virtual objects for the urban lifeline. large urban lifeline infrastructure’s service state will influence the operational efficiency of the city. Hence, when and which parts of it should be maintained are pieces of public safety information that must always be available.

For a urban lifeline infrastructure with complex environments, loads, and structure types, regular manual or low-intelligence equipment inspections are time-consuming and labor-intensive and lack the ability to quickly provide feedback and synchronize information.

Digital twins (DTs), as a new means of infrastructure informatization, digitization and intelligence that emerged after building information modeling, emphasize the real-time nature of twin modeling more than BIM does and can better fuse multi-source and multimodal observation data. Hence, it has potential to establish an effective, economized and intelligent DT system to closely link the virtual model of data integration with the physical entities of infrastructures. This system will enable predictive diagnosis of hidden risks within infrastructure entities based on DT-driven deduction.

To achieve this goal, three main bottleneck issues must be solved.

(a) Seeking a balance between effectiveness and cost for the observation of large infrastructures under the constraints of limited sensing ability.

(b) Dynamic modeling of time-varying states and structural effects for physical entities during the operation of lifeline infrastructures.

(c) Representation of multi-objective performance and predictive diagnosis of hidden risks for high-redundancy systems of lifeline infrastructures.

Addressing three main bottleneck issues, Associate Professor Hanwei Zhao et al. from Southeast University proposed four steps of researches strategies mainly regarding the point monitoring–area detection fusion of observations and the use of DTs to diagnosis infrastructure deterioration of safety, durability, and availability:

First, theories should be developed for the intelligent processing of point monitoring data on structural actions and effects, as well as for optimizing measurement points.

Second, breakthrough technologies are needed for intelligently processing the area detection data on the distribution of structural effects and for updating the data through fusion.

Third, dynamic evolution systems of DTs for lifeline infrastructures are expected to be established via multi-source big data observation.

Fourth, predictive diagnostic approaches for identifying hidden risks of lifeline infrastructures based on the dynamic DT system must be proposed.

A complete process from DTs to virtual simulation to virtual and real symbiosis for the full-field effects of lifeline infrastructures can be performed after these steps are completed. The complete process can help break the information barrier between infrastructure entities and DT systems. Then, monitoring and detection data can be converted into diagnostic indicators of structures about hidden risks, which will support the intelligent operation and maintenance of future cities.

This paper “Predictive diagnosis of hidden risk for urban lifeline infrastructures driven by digital twin modeling of multisource observations: perspective” was published in Smart Construction.

Zhao H, Zhang X, Ding Y, Guo T, Li A, et al. Predictive diagnosis of hidden risk for urban lifeline infrastructures driven by digital twin modeling of multisource observations: perspective. Smart Constr. 2024(3):0014, https://doi.org/10.55092/sc20240014.