In the following decades, the value of information has been paid more and more attention and become the main wealth of society. As a tool to tap the value of information, information technology is developing with each passing day. Information and information technology have changed everyone's work and lifestyle, promoted the rapid economic development and brought subversive changes to the whole society.
After entering the 2 1 century, the development of information technology has undergone new changes.
The evolution of computing technology represented by cloud computing, big data and artificial intelligence, and the leap of connection technology represented by all-optical network, 4G/5G and Wi-Fi 6 make people put forward higher expectations for digital technology. People hope to further realize digitalization, networking and intelligence on the basis of informatization, and transfer the surging digital kinetic energy from the personal consumption field to various vertical industries including industrial manufacturing, transportation and logistics, education and medical care, so as to realize the digital transformation of the whole industry and society.
In other words, digital technology not only helps consumers to socialize and entertain better, but also helps enterprises to upgrade their manufacturing processes, improve their business processes and further increase their productivity. In addition, it is necessary to help the government improve its governance capacity, optimize management efficiency and improve the quality of urban life of residents.
Digital twin technology was born under this background.
20 1 1 in March, the United States air force laboratory first proposed digital twinning. At that time, they applied this concept to the digitalization of fighter maintenance.
Soon after, the other two companies paid attention to the digital twin and decided to carry it forward in the civil field. The two companies are General Electric (GE) in America and Siemens in Germany.
General Electric and Siemens are world-class industrial giants. They have long been concerned about industrial automation and digital transformation, and have also been studying Industry 4.0.
For them, the product of the development of digital twin information technology to a new stage is a typical industrial digital technology, which represents the future direction of the deep integration of industrial manufacturing means and digital technology. To this end, they have invested a lot of resources in the research and development of digital twin technology and pushed it to various fields around the world.
After talking for a long time, what exactly is digital twins?
The official concept of number pairing is hard to understand. It goes like this:
Digital twinning refers to the comprehensive use of information technologies such as perception, calculation and modeling, and the description, diagnosis, prediction and decision-making of physical space through software definition, so as to realize the interactive mapping between physical space and cyberspace (which can be understood as digital virtual space).
Deleting the description and refining the backbone will become simpler: "Digital twinning is the interactive mapping between physical space and digital virtual space."
To put it more simply, digital twin is to create a digital version of "clone" based on a device or system. This kind of "cloning" is also called "digital twin".
For many people, digital twins are easily confused with "digital modeling". After all, digital modeling is also a simulation clone.
But in fact, digital twins are very different from "digital modeling". The characteristics of digital twins can be summarized in four words-dynamic, full life cycle, real-time/quasi-real-time, and bidirectional.
The so-called "dynamic" means that the real-time state of the ontology and the state of the external environment will reappear on the digital twin through sensors and other means. In other words, twins are not static, but changing.
"Full life cycle" refers to the whole life cycle of digital twins, including design, development, manufacturing, service, maintenance and even scrapping and recycling. Not only help enterprises to create product ontology, but also help enterprises to use and maintain ontology.
"Real-time/quasi-real-time" is easy to understand, that is, the "dynamic" digital response mentioned above is realized in real-time/quasi-real-time, and there is no big delay and obvious lag.
The characteristic of "two-way" is very critical. Traditional modeling is often one-way-build a model, and then make ontology according to the model. Digital twins are completely different. In addition to receiving ontology data, twins can also send data to ontology in the opposite direction. Enterprises can take further actions and interventions on ontology according to the information fed back by twins.
Technically, the digital twin technology system is very huge. The process of perception, calculation and modeling covers many technical fields, such as perception control, data integration, model construction, model interoperability, business integration, human-computer interaction and so on, and the threshold is very high.
The technological competition of digital twins is actually a game of the comprehensive strength of advanced ICT technologies such as cloud computing, big data, 3D modeling, industrial Internet and artificial intelligence.
Digital twinning is essentially a technology to simulate the real ontology of physical space with the help of digital space twinning model. The reason for simulation is nothing more than two reasons: first, the cost of physical ontology is expensive, and the cost of trial and error is too high, which is beyond the affordability. Second, the physical ontology is unique, does not support physical replication, and has no chance of trial and error.
As mentioned above, the United States Air Force Laboratory and General Electric Company, the earliest digital twins, are expensive airplanes and airplane engines. By 20 18, GE has developed 1.2 million digital twins. According to their own account, they created a digital twin for every engine, every turbine and every MRI.
After establishing digital twinning, they collected the running data of physical ontology and put it on twinning. Then, they can boldly innovate, make full trial and error, make product design changes, conduct simulation tests, and observe the effect, so as to judge whether to implement actual product changes. In this way, the cost and risk of trial and error are greatly reduced and the product development cycle is shortened.
What kind of system is unique and does not support physical replication?
Of course, it is the kind of large-scale, real, open and in-use system. Everyone should have thought that the city we live in every day is such a system.
Cities are extremely complicated. In cities, there are millions or even tens of millions of people, countless buildings and vehicles, and intricate infrastructure networks (roads, water, electricity, gas, communications). We can't do experiments directly in the city, nor can we copy a physical city to do experiments. Therefore, we need to build a virtual city in digital space with the help of digital twin technology, and carry out simulation, experiment and trial and error to improve the management and operation efficiency of the city.
Traffic is one of the most important functions of a city. Take Tencent's digital twin platform as an example to learn more about how digital twin technology can empower the application of smart transportation industry.
First of all, let's look at the construction of urban traffic digital twins.
Tencent can truly restore the main elements of the whole city, such as buildings and roads, based on its own high-precision map data by using city-level 3D reconstruction technology. Then, there are static elements such as tree greening, bus stops, traffic signs and traffic markings.
With static elements, what about dynamic elements such as vehicles and pedestrians? Is it generated at any time like in computer games?
Of course not. Vehicles and pedestrians in the traffic twin system are not completely virtual, nor are they randomly generated.
Digital dual platform collects real-time images of road monitoring through road test cameras, and detects and senses various targets (vehicles, driving, etc. ), then "extract" the target data, model it and integrate it into the digital twin environment synchronously.
Only in this way can we truly realize the deep integration of real road traffic and virtual environment. Real-time traffic flow big data has been truly restored in digital space.
Under the large-scale lane-level real-time simulation in the whole city, the traffic flow data can be presented intuitively in color (traffic heat map) just like we use the map navigation APP. The degree of urban traffic congestion is clear at a glance.
After the construction of traffic digital twins is completed, from God's point of view, we can do many things.
Firstly, the change of traffic flow under special circumstances can be deduced. Simulate traffic accidents, or hold large-scale activities or performances, observe the changes of traffic flow, test the carrying capacity of traffic networks, and provide decision-making basis for urban traffic authorities to formulate emergency plans.
Secondly, it can simulate the design optimization scheme of traffic lights and provide technical support for the management department to optimize traffic management scheduling.
In addition, based on the virtual environment, it provides dynamic route planning for emergency vehicles such as ambulances and fire engines, saving lives.
In the digital twin environment, Tencent also introduced game engine related technologies to freely simulate various weather conditions, evaluate the impact of weather on the transportation capacity of the transportation network system, and make emergency plans in advance in disaster weather environment.
It is worth mentioning that the traffic digital twin platform has not been closed. It is possible to open low-cost and low-threshold API interfaces to partners, which is convenient for them to integrate business systems and make the transportation industry bigger and stronger.
In addition to urban traffic, intercity expressway is also an important application scenario of traffic digital twins.
In this scenario, digital twin technology emphasizes active operation, that is, managing the whole expressway.
Data transmission can be realized by ETC, camera, vehicle networking terminal, RSU (roadside unit), cellular base station and even satellite. Through the cloud or on-site MEC (Edge Computing Node), data can be calculated and processed.
For expressway management departments, based on the perceived traffic flow, road surface, weather and accident data, it is easy to realize the active and safe fine control of expressway elements such as lane level positioning and guidance, dynamic route arrangement and traffic infrastructure regulation.
For drivers, traffic digital twin technology has the accurate perception ability with an accuracy rate of 95% and the communication ability with an end-to-end delay of 300 ms. It can send abnormal events to the driver's dual terminals in the form of vision and sound, thus improving the driver's safe traffic ability.
Especially in bad weather (such as fog, rainstorm, etc.). And at night, the driver's line of sight is blocked, and real-time and accurate surrounding road conditions can be obtained through double driving.
Based on traffic digital twin technology, the system can also provide LBS companion service for car owners, and push services to car owners through WeChat and map App to improve driving experience.
Another remarkable feature of traffic digital twin technology is computability.
It is mainly reflected in the computability of road traffic and the spatial computability of simulation and prediction. Through simulation and prediction, the future traffic flow of 1 hour can be predicted, which supports the simulation of emergency plans in different traffic management scenarios and provides scientific and quantitative decision-making basis and data support for decision makers.
Finally, let me talk about the unmanned driving that is very hot now.
Driverless driving is an ultimate development direction of intelligent travel. At present, major manufacturers are actively conducting research and testing on related technologies.
However, to achieve the popularization of driverless technology, the most important point is the massive testing and learning of driving data. However, the current laws and regulations, as well as road conditions, do not allow unmanned vehicles to enter the real road environment for testing at will. At this time, the digital twin environment can play a role.
Based on the digital twin environment, the automatic driving test environment can be freely constructed, and the parallel accelerated test of the city-level cloud simulation environment can be realized in the cloud, and the test is 7×24 hours, and it can test 6.5438+million kilometers every day. This is simply the blessing of driverless technology. It can greatly shorten the development cycle of driverless technology and accelerate the popularization.
In a word, digital twinning, as a digital conversion technology of "combination of reality and reality", is accelerating in various fields. The rapid development of industrial Internet has promoted its value explosion.
In the next century, if human beings want to achieve greater ambitions, virtual space technology represented by digital twins will be an important tool and scene.
What is the potential of digital twins? Let's wait and see!