Risk Twin: Real-time Risk Visualization and Control for Structural Systems

Zeyu Wang,Ziqi Wang

2024-03-01

Abstract:Digital twinning in structural engineering is a rapidly evolving technology that aims to eliminate the gap between physical systems and their digital models through real-time sensing, visualization, and control techniques. Although digital twins can offer dynamic insights into physical systems, their accuracy is inevitably compromised by uncertainties in sensing, modeling, simulation, and controlling. This paper proposes a specialized digital twin formulation, named Risk Twin, designed for real-time risk visualization and risk-informed control of structural systems. Integrating structural reliability and Bayesian inference methods with digital twining techniques, Risk Twin can analyze and visualize the reliability indices for structural components in real-time. To facilitate real-time inference and reliability updating, a "simulation-free" scheme is proposed, leveraging precomputed quantities prepared during an offline phase for rapid inference in the online phase. Proof-of-concept numerical and real-world Risk Twins are constructed to showcase the proposed concepts.

Applications

What problem does this paper attempt to address?

The paper attempts to address the problem of achieving real-time risk visualization and risk-based control in structural systems. Specifically, the paper proposes a digital twin technology called "Risk Twin," which aims to bridge the gap between physical systems and digital models through real-time sensing, visualization, and control methods, thereby enhancing the safety of structural systems.

### Main Issues
1. **Uncertainty Issue**: Existing digital twin technologies have uncertainties in the sensing, modeling, simulation, and control modules, which affect the accuracy of the digital twin models.
2. **Risk Assessment Issue**: There is a lack of a unified risk assessment framework that can quantify and compare the risks of different structural components.
3. **Real-time Control Issue**: There is a need for a method that can achieve real-time risk visualization and risk-based control to improve the safety and reliability of structural systems.

### Solutions
1. **Risk Twin Framework**: A digital twin technology specifically for real-time risk visualization and risk-based control is proposed, combining structural reliability and Bayesian inference methods.
2. **"No Simulation" Scheme**: To achieve real-time inference and reliability updates, a "no simulation" scheme is proposed, utilizing precomputed data from the offline stage for rapid inference in the online stage.
3. **Practical Application**: A proof-of-concept numerical and practical Risk Twin system is constructed to demonstrate the effectiveness of the proposed concept.

### Key Technologies
1. **Bayesian Inference**: Used to statistically update the posterior distribution of basic random variables, thereby updating the digital model.
2. **Risk Shadow**: A digital representation of risk used to display the reliability index of structural components in real-time.
3. **Bidirectional Interaction**: Bidirectional interaction between the physical system and the digital model, where sensor data updates the digital model, and adjustments in the digital model also affect the physical system.

### Experimental Validation
1. **Numerical Experiment**: Demonstrates the application of Risk Twin on a simple support plate, validating the effectiveness of real-time Bayesian inference.
2. **Practical Experiment**: Demonstrates the application of Risk Twin in the control of a cantilever beam by a robotic arm, validating the effectiveness of risk shadow and human-machine interaction.

### Conclusion
By introducing the Risk Twin framework, the paper successfully addresses the problem of real-time risk visualization and risk-based control in structural systems, enhancing the safety and reliability of structural systems.