Is there really another "me" in the world? Digital twin: I am

Produced by | Science Popularization China

Author | Salted Fish in the Sea

Producer | China Science Expo

Have you heard the story of the "Three Bian Que Brothers"?

It is said that King Wen of Wei once asked the famous doctor Bian Que for advice: "You have three brothers, all of whom are good at medicine. Who is the best?" Bian Que said: "The eldest brother is the best, the second brother is not so good, and I am the worst among the three." King Wen of Wei was puzzled, and Bian Que said: "The eldest brother treats the disease by giving medicine to eradicate the root cause before the disease occurs; the second brother treats the symptoms before the disease just appears; I treat the disease very seriously by using major surgery or using poison to treat poison to relieve the patient's condition or cure it quickly, so I am famous all over the world, but in fact my medical skills are the worst."

This principle is also universal in the industrial world, that is, eliminating hidden dangers before an accident occurs can maximize the protection of people's lives and property.

There is a famous Heinrich's Law in the industrial world. Heinrich, an industrial safety engineer, mentioned in his book "Industrial Accident Prevention": Behind one major safety accident, there must be 29 minor accidents and 300 potential hidden dangers. If you seize the opportunity and eliminate unsafe factors in time before an accident occurs, you can avoid many major casualties.

So who can solve hidden dangers before accidents happen? Digital twins.

What is a digital twin?

One of the purposes of digital twin technology is to prevent accidents before they happen.

In 2012, NASA gave a conceptual description of digital twins: Digital twins refer to the integration of multidisciplinary and multi-scale simulation processes that make full use of physical models, sensors, operating history and other data. It serves as a mirror of physical products in virtual space, reflecting the entire life cycle of the corresponding physical products.

This concept may seem complex, but it will become clear if we look at its application.

Aircraft engines are important components on aircraft, and their stability and normal operation are extremely important to ensure flight safety. In order to effectively detect the operating status of the engine, aircraft engine manufacturers have created digital twins for it.

Engineers first create an exact virtual replica of the engine in a computer. Then they install various sensors in the real world to collect data, which is then transmitted in real time to the digital twin in the computer.

By running the engine in the virtual world, the real-world situation can be simulated. After a long period of virtual operation, the staff can collect information about the engine's operation and understand when it needs maintenance.

The engine is being tested. It is impossible to simulate all situations in the real world, but this is not a problem in the computer. Image source: The US National Archives

In this way, we can actually perform preventive maintenance on aircraft engines, thereby greatly reducing the regular downtime of aircraft maintenance.

Furthermore, in the computer, engineers can test the operation of the engine under various extreme and unexpected conditions, which may be difficult to simulate in the real world. This can deepen the understanding of the machine and improve its reliability under various conditions.

Where do digital twins come from?

Before the emergence of digital twins, the technology for modeling and analyzing real-world objects in computers had already existed and had become very mature and widely used in various fields.

Digital twins are widely used in various fields. The picture shows the digital twin of an offshore oil platform. Image source: Wikipedia-SumitAwinash

For example, in the field of engineering, if you want to build a bridge, how should the bridge be designed to achieve the strongest load-bearing effect? ​​First, build a model in the simulation software, continuously optimize the model, and then build the bridge in reality based on the optimized model.

Computer simulation technology is the basis for the emergence of digital twin technology. The difference between digital twins and simulation models is that digital twins are evolutionary. Digital twins will continuously receive various types of information from actual objects, adjust their status in real time, and strive to achieve a state that corresponds to the actual object in real time.

Based on the information provided by the digital twin, the computer issues instructions to adjust the operating parameters of the physical world machine in real time to achieve the best operating state. Image source: Deloitte University Press

Let's take another example that is more relevant to our lives. The engine tachometer on every car dashboard has a red zone, which means that if the engine speed reaches this range, there is a high probability of a problem. This red zone remains unchanged throughout the life of the car.

But the reality is that the usage of each car is different. Some cars are well maintained and there is no problem when the engine speed reaches the red zone; some cars are poorly used and have problems before the engine speed reaches the red zone.

If the engine in a car can have a digital twin that runs synchronously with the engine in the virtual world, then the engine's red zone can be changed in real time based on the operation of the digital twin, greatly reducing the possibility of engine failure.

David Gelernter foresaw the concept of digital twins in his 1991 book Mirror Worlds. In 2011, the U.S. Air Force Research Laboratory first explicitly mentioned digital twins, hoping to digitize fighter maintenance, and digital twins were their proposed solution.

The U.S. Air Force has deployed a large number of sensors on the F35 fighter jets, and transmitted the data collected by the sensors to a pre-established digital model. The model will change in real time with the data, accurately simulating the actual status of the fighter jet.

The F35 is a new generation of fighter jets that is now widely used in the world. Image source: Israeli Air Force

When General Electric helped the Air Force develop the digital twin of the F35 fighter, it also became very interested in this concept and continued to invest in research. As of 2018, General Electric claimed that they had created a corresponding digital twin for every engine, turbine, and MRI they produced.

The rapid development of computers has enabled us to process massive amounts of data from physical entities. The emergence of machine learning and AI has made it possible to build extremely complex virtual models. The massive application of sensors, the establishment of high-precision digital models, and the data interaction between physical entities and virtual models are the basis for the existence of digital twins.

Will we have our own digital twin?

Every person is unique. Unlike products produced by modern manufacturing assembly lines, humans are extremely complex. With the development of technology, the concept of smart healthcare has been proposed and received more and more attention.

In recent years, smart watches have appeared on people's wrists. These watches can detect basic information such as heart rate and oxygen content in the blood. If we sit for too long, the watch will remind us to get up and walk around. When we exercise, the watch will also issue an early warning if our heart rate is too fast.

Now, with the help of machine learning, computers can judge our mood through pictures of faces; during the process of losing weight, we can know how many calories a food contains by taking a photo with a mobile phone; smart watches will record our sleep data from the previous night.

In a sense, this is the most basic version of the human digital twin.

Smart watches and other wearable smart devices can collect a lot of data about the human body. Image source: pixabay

However, for the huge and complex human system, only the above-mentioned data cannot accurately predict the human body's condition. In order to obtain a more accurate digital twin, a large amount of data needs to be collected for analysis. This includes the air quality of the human body, lighting conditions, vaccines administered to the human body, electroencephalograms, electrocardiograms, the types and quantities of food consumed, and so on.

Creating a digital twin that can grow in sync with a real person remains an impossible task today.

Despite this, people are still confident about the bright prospects of digital twins in the field of medical health. In 2014, a digital twin development project called the "Live Heart Project" was launched. Experts from all walks of life and medical practitioners hope to create a digital twin of the human heart.

A few years later, the project delivered the first reference model in which doctors can reproduce any cardiovascular disease and safely test treatment options, which doctors in some hospitals use to plan personalized heart surgery procedures for different patients.

Some emerging medical methods may cause great harm when tested in reality, but they can be avoided in digital twins.

First, a general model is established. Based on this model, a corresponding personalized model is established according to the information of different patients. Image source: self-made by the author

As COVID-19 sweeps the world, researchers have launched a new digital twin project, the Living Lung Project (LLP), to understand the long-term effects of the virus on human lungs. In virtual space and time, we can speed up the flow of time and analyze the long-term effects of the disease.

Image source: Wikipedia

What was once thought to be impossible to build a digital twin of a complete passenger aircraft is now commonplace for aircraft manufacturers. The construction of a digital twin of the human body still presents many unique challenges today, but people believe that the speed of scientific and technological progress will overcome these challenges.

Digital twin is a complex field that crosses multiple disciplines. Mathematics, physics, materials science, computer science, and biology are all essential disciplines. More and more researchers are entering the field of digital twin research, and digital twins are beginning to appear in urban planning, smart factories, biopharmaceuticals and other fields. Another "me" may be in the near future.

References

[1] Wei S. Is Human Digital Twin Possible?[J]. 2021.

[2] Lv Q, Zhang R, X Sun, et al. A Digital Twin-Driven Human-Robot Collaborative Assembly Approach in the Wake of COVID-19[J]. Journal of Manufacturing Systems, 2021(12).

[3] Koen B, Filippo SDS, Jeroen VDH. Digital Twins in Health Care: Ethical Implications of an Emerging Engineering Paradigm[J]. Frontiers in Genetics, 2018, 9:31-.

[4] Barricelli BR , Casiraghi E , J Gliozzo, et al. Human Digital Twin for Fitness Management[J]. IEEE Access, 2020, 8:26637-26664.

[5] Glaessgen E , Stargel D . The Digital Twin Paradigm for Future NASA and US Air Force Vehicles[C]// Aiaa/asme/asce/ahs/asc Structures, Structural Dynamics & Materials Conference Aiaa/asme/ahs Adaptive Structures Conference Aiaa. 2012.

[6] Tao Fei, Liu Weiran, Liu Jianhua, et al. Digital twin and its application exploration[J]. Computer Integrated Manufacturing Systems, 2018, 24(1):18.

[7] Tao Fei, Zhang He, Qi Qinglin, et al. Theory and application of digital twin model construction[J]. Computer Integrated Manufacturing Systems, 2021, 27(1):15.

[8]https://zhuanlan.zhihu.com/p/99249900

[9]https://en.wikipedia.org/wiki/Digital_twin

[10]https://www.lockheedmartin.com/

[11]https://www.ge.com/digital/applications/digital-twin

Special Tips

1. Go to the "Featured Column" at the bottom of the menu of the "Fanpu" WeChat public account to read a series of popular science articles on different topics.

2. Fanpu provides a function to search articles by month. Follow the official account and reply with the four-digit year + month, such as "1903", to get the article index for March 2019, and so on.

Copyright statement: Personal forwarding is welcome. Any form of media or organization is not allowed to reprint or excerpt without authorization. For reprint authorization, please contact the backstage of the "Fanpu" WeChat public account.