There has been a surge in digital twin related research over the past 2-3 years, however, the term has become a buzzy word that academics and industry overuse. In particular, a couple of tech giants have used Digital Twin for marketing their existing products&services. This chaotic situation has made it challenging to understand the core research questions on digital twin.
Is digital twin a concept that only targets engineering application innovation with existing technologies?
OR, there are unique research questions for digital twin itself?
Very good question! The situation of Digital Twin or even Intelligent Manufacturing should be changed. At this stage, related researches should be problem-driven rather than concept-driven.
There is no doubt that AI is starting to disrupt the workplace through digital transformation, resulting in extensive use of the digital twin. This "digital twin" is effectively a virtual representation of a physical object or system. As it has evolved, it has come to also encompass larger entities such as buildings, factories, and cities. It includes IOT data, advanced computer systems, digital processes, electronic documents and advanced analytics which all model physical space.
The combination of AI with the digital twin results in significantly enhanced productivity. This is not theory; this is a fact and is quantifiable. AI enhances workforce productivity and improves safety, reliability, quality and security. Through efficiency gains and reduced waste, AI is creating an overall greener environment with enhanced sustainability. AI also helps the workers themselves–by upskilling them and allowing staff to make more accurate data driven decisions. Studies show that there is not enough new qualified staff to replace the knowledge of an aging workforce rapidly approaching retirement. AI also helps to facilitate and reduce this gap...
https://www.automation.com/en-us/articles/september-2020/analytics-ai-processes-revolutionize-industrial
Agree with Yuqian Lu that DTs have become a meme. It seems we, as a community of researchers, can hardly influence this. But we have other tasks.
- It is necessary to outline the limits of this phenomenon. These limits are unnecessarily expanded, which is partly due to the ambiguity and richness of the metaphor proposed by Michael Grieves. But with all this diversity, the key identifying feature of DT remains – they must remain in “hot mode” relative to the original; moreover, the synchronization rate of the original and the twin must be commensurate with the object control rate which can be very fast or very slow. That is, if a model is updated significantly less often than an object is controlled, then this is not DT. Of course, it makes no sense to update DT significantly more often than the control signal changes.
- R&D of tools and techniques used in DT also remains in our responsibility area - for example, the development of methods for constructing physical object hybrid models that combine the advantages of First Principles Modeling and Machine Learning. This is important in equipment predictive analytics, where there is no alternative to DT. Increasingly widespread DT-platforms require unification of their communication, computing and infrastructure characteristics. The corresponding theoretical basis is still far from being completed.
This might help you.
Understanding The Value Of Digital Twins: Apply Advanced Analytics To Move To Digital Reliability
What does practical digital reliability look like as we seek to successfully navigate the gauntlet of digital transformation? First, we need to focus on creating real solutions and technologies that facilitate a move forward on digital reliability and truly make a difference to the operators on the ground. The digital twin—along with its enabling technologies of Industrial AI and machine learning—can provide companies with the tools they need to create the digitally-enabled organization that can thrive in today’s volatile economy...
https://info.smartindustry.com/en/webinars-2021-portal-insight-understanding-the-value-of-digital-twins-apply-advanced-analytics-to-move-to-digital-reliability_ce_digital_mfg?utm_campaign=SI_2021_CAMP_SIEnews&utm_medium=email&_hsmi=150936010&_hsenc=p2ANqtz--sLvgMPybNQ60ro36_y91tPE84AwFluUqX9JyPdKXp3PoxO3NpiHBNhZkkdBRX4XBm7KHoNqKf8qvbQlHccKXBvq0FmA&utm_content=150936010&utm_source=hs_email
This white paper describes how Siemens Digital Industries Software’s technology standardizes data formats to enable the development of a comprehensive digital twin for design, process and production, allowing manufacturers to simulate changes in a safe environment without affecting real-life processes or production...
Intelligent data exchange provides the flexibility needed in Industry 4.0 factories. As manufacturing becomes smarter, the data architecture must enable connected and collaborative communication throughout the evolving manufacturing ecosystem. Intelligent data exchange also contributes to ongoing improvement by closing the loop between the virtual (digital twin) and actual product...
Really interesting question! I'm personally quite interested in the topic so I'm looking forward to additional replies. However, in regards to my own input, I don't think the concept of digital twins is exclusive to engineering applications, nor is it exclusive to existing technologies.
Many companies across various industries use the technology behind digital twins (and I agree, this has definitely grown to be quite "buzz wordy") in order to have a software or mock-version of a manual process and it's much easier to optimize it using modeling or running tests on a program. Moreover, it's used in many industries to prototype and mockup new innovations and processes, not only those that exist (common in aerospace engineering, automotive, biomedical).
To the attention of Monica S Spychala - please, see Discussion "Nuances in the digital twin concept. Opinions?".
https://www.researchgate.net/post/Nuances_in_the_digital_twin_concept_Opinions?_ec=topicPostOverviewFollowedQuestions&_sg=yIYME3VeU_QPDtOkIHIythLmX6HbOwgrQM1mA5IZFunOuSuTj7cI_AOpvOfok_3srCoD98_MyQMW96AC
Building information modeling, digital twins create green, smart cement factory
Pictures and models have been used to design every type of costly and complex construction throughout human history. More recently, increasingly dynamic simulations and digital twin software are processing data close enough to real-time to also optimize assets and operations...
By using Bentley’s BIM applications in a connected data environment with its iTwin cloud-based processing services that display iModels from the original design applications, Citic also established a digital twin to visually manage equipment and integrate process data related to equipment operations at Fujian Ansha. This process facilitates online, intelligent equipment monitoring and diagnostic services, and establishes a preventive maintenance system that enhances operational safety and improves equipment reliability. Through Bentley’s integrated, cloud-based applications, Citic produced accurate 3D models of the cement plant's main equipment, and can now digitally monitor plant equipment and processes, achieving whole lifecycle BIM management from design through delivery and operations...
https://www.controlglobal.com/articles/2021/building-information-modeling-digital-twins-create-green-smart-cement-factory?utm_campaign=CG_2021_Newsletters&utm_source=hs_email&utm_medium=email&utm_content=156875368&_hsenc=p2ANqtz--YxdOEXGJ2u9vod36KRTiHSQ6wiwVKh1HKQ08x11rIPMjs7DtFf--b-VoQXObyMRVyes7_qS0DR1k48TWhYHFgItg14Q
Fine example of the application of digital twin.
Digital Twins Enable the Autonomous Paper Mill
Robust and well-maintained control system optimization, advanced process control, and model-based predictive control can increase production, decrease costs, and autonomously operate the mill of the future...
https://www.automation.com/en-us/articles/september-2021/digital-twins-enable-autonomous-paper-mill
Hi Dr Yuqian Lu . I hope the following link could help: https://hal.archives-ouvertes.fr/hal-03218786/document
You can refer to following links
https://resources.sw.siemens.com/en-US/white-paper-what-is-a-comprehensive-digital-twin?stc=indi100002&ef_id=CjwKCAjw7fuJBhBdEiwA2lLMYcNYgUqMNqx6JoILMaR2K_KmIB7I1yggqM1mSeNGYDat26aSz35P_RoCfS4QAvD_BwE%3AG%3As&s_kwcid=AL%21463%213%21544135253669%21e%21%21g%21%21what+is+digital+twin+in+manufacturing
https://www.asme.org/topics-resources/content/7-digital-twin-applications-for-manufacturing
https://slcontrols.com/en/what-is-digital-twin-technology-and-how-can-it-benefit-manufacturing/
https://www.biz4intellia.com/blog/rise-of-digital-twin-in-manufacturing-sector/
To me, the key potential of the digital twin concept is the interaction between the real asset and the corresponding numerical representative model, through available measurement data. Specifically, the available measurement data can benefit to improve the numerical representative model, and in return, such an improved model can improve the prediction accuracy of the real asset behavior.
Without such ability, the digital twin will only provide value as a fancy (maybe even not fancy at all) visualization tool, benefiting engineers to understand the asset and react more quickly for potential operations. The interaction between real assets and the numerical model is mainly realized by techniques of system identification and state estimation.
I recommend the following survey paper published in 2020. For me, it clears out several key issues.
Title: Digital Twin in the IoT Context: A Survey on Technical Features, Scenarios, and Architectural Models.
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9120192
Open-source collaboration drives digital twin innovation...
"As a consortium, our collective goal is to progress digital twin technology out of the lab and into the marketplace. Digital twins can be difficult to apply across the entire life cycle, where efficiency is often stifled by data silos and a lack of interoperability. Integration with legacy environments can also be challenging," said Dan Isaacs, CTO Digital Twin Consortium. "Digital Twin Consortium's open-source community will speed the cross-industry adoption of digital twin technologies and techniques and drive digital twin technology toward interoperability and market readiness."
https://www.automation.com/en-us/articles/october-2021/digital-twin-consortium-open-source-collaboration
This article is the first in a three-part series seeking to clarify what a digital twin means in the chemical process industries, how models and twins are used, and some techniques that can convert a model to a twin...
https://www.controlglobal.com/articles/2021/understanding-the-digital-twin-part-1?utm_campaign=CG_2021_Newsletters&utm_source=hs_email&utm_medium=email&utm_content=179271426&_hsenc=p2ANqtz-_YnO1THFYycuGMayzJFDfKmsenXGdQ7jv8HRETWaeozzmkZDjQyCH37yuNqx6zjCg_Dzt8vX1qR-5PDoofuXYPk2eRKg
A Digital Twin is a digital representation of something (mostly) physical and it enables exploration and verification. A 3D representation of a building can make it possible to virtually explore being inside it or can be used more conventionally to verify conformance to specifications or compatibility to other building parts. A digital representation of the surrounding of a car and the physical parts of the car can make it possible to test the autonomous SW. A digital representation of a cyber-physical system like a car, a ship, a plane, with operational data can make it possible to explore, learn and improve.
In general digital representation can remove some of the limitations of needing expensive, slow, unsafe physical or real stuff and increase the speed or possibility of exploration.
Trust in the simulation is key as we would need to trust that the representation has the needed accuracy and properly copies the real thing for our purpose. The digital twin produces synthetic data that needs to properly represent the real thing or it will invalidate the results and conclusions made upon the digital twin.
Accuracy or the fidelity of models in copying the details of the reality, also comes with a cost, in terms of development cost, but not to forget the computational cost (computing time or power). Therefore, accuracy should often in practice be at a minimum required level.
I think there are several research questions related digital twins:
1. What different applications or types of use can digital twins support?
2. What are important aspects of digital twins in order to trust the decisions made upon using them?
In my work, we both explore the potential use of digital twins, but primarily we are concerned on what grounds we should trust the simulations or decisions made upon them.
How to adapt the model?
The digital twin begins life as a prototype model, intended to match the process, but possessing ideal attributes that need to be adapted to the actual process. First, one must adjust model coefficients and constituent relations to best match the model to actual process operation. Second, over time as the process is reconfigured or substantially altered, reflect those process changes in the model. This may also require adjustment of model coefficients. Third, as the process operates, there will be continual changes in features such as ambient heat losses, heat exchanger fouling, catalyst reactivity, friction losses in piping, raw material properties, etc. Incrementally adapt the model to match the local reality...
https://www.controlglobal.com/articles/2021/understanding-the-digital-twin-part-2?utm_campaign=CG_2021_Newsletters&utm_source=hs_email&utm_medium=email&utm_content=192408810&_hsenc=p2ANqtz--lgcTGvjA_XDQDdw12Egn7PWg016tdJeJm_ihYjxdnhgb1d8S8ESmZgVbKDzUA3-d2mUyUWep43uCU5e2FTevD1FYBEkK3irwcgw2Kzi92iQMli6A
The NSW Spatial Digital Twin was an evolutionary leap when it came to shaping communities and boosting productivity allowing customers, developers and planners to see into the future...
https://pacetoday.com.au/nsw-spatial-digital-twin-evolves-planning-data/
The digital twin methodology provides precise insight for optimizing parameters to improve KPIs. Expanding on this concept, a CLDT creates an accurate replica of assets’ current states to forecast accuracy, beyond that of a standard digital twin without feedback. CLDTs compare current state conditions with numerous adaptations to determine an optimal future state, aided by AI/ML...
https://www.automation.com/en-us/articles/november-2021/closed-loop-digital-twins-holiday-shipping
Explore the possibilities of integrated product and production lifecycles with the comprehensive DigitalTwin. Combine the real and digital worlds and leverage the infinite amount of data to use your resources more efficiently. The goal is to do more with less, making industry more sustainable...
https://new.siemens.com/global/en/company/topic-areas/digital-enterprise.html?linkId=300000001836237
https://www.linkedin.com/in/ljubomirjacic/detail/recent-activity/
With a simulation in the cloud on an open Internet of Things (IoT) cloud platform, users can map data from the plant floor to the digital twin model. This data is used to create insights for optimizing conditions and control variables on the production lines to maximize throughput and other KPIs. Some cloud platforms include a dedicated app for preparing and aggregating time series data into a simulation application.
An open IoT cloud platform empowers users to calibrate their digital twin with historical production data...
https://www.automation.com/en-us/articles/january-2022/digital-twins-calibrated-operational-data-drive
While model coefficients should be adjusted, noise can cause undesirable effects...
When a digital twin’s model coefficient values are adjusted in real-time, noise, digitization, spurious signals and other data features can lead to unwarranted changes in model coefficient values. This is termed tampering...
Tampering increases variability, costs energy, increases wear and adds confusion...
Prevent process noise from continually perturbing KPI values. Rather than using the conventional first-order filter, consider SPC technique to prevent tampering...
https://www.controlglobal.com/articles/2022/understanding-the-digital-twin-part-3?utm_campaign=CG_2021_Newsletters&utm_source=hs_email&utm_medium=email&utm_content=200259296&_hsenc=p2ANqtz-9NSMDHYVsf7RdCC2uwc1QD58Cre7z1bVN9isZxhHiWl-l9HLfTtWDJMIaLUm9MQx_g51T0DmuwsA6aLvYg7K1LtDEGtoXHEQU4dkIswu0JhYElPZw
I have compiled some questions about the Digital Twin that have arisen to me in the following question:
https://www.researchgate.net/post/Digital_Twin_doubts_Any_help
In my opinion the Digital Twin is nothing more than a notion by which virtual replicas of real world entities are created whose purposes vary according to the research objectives and interests pursued. It is not at all a novel technology, but, as I say, a feasible notion based on the new technologies that have been developed in recent years. In my opinion, the expansion of the concept to different sectors and the adoption of the concept by each one "in their own way" make it very difficult to achieve a unified vision of it today. Perhaps it will be achieved by sector, as in the manufacturing industry, where it actually has a wide application (or at least it is very mentioned... although many times as a buzzword, being the underlying applications simply deployed ML models giving real-time predictions...). The international organizations dedicated to its standardization have a great task ahead... All my respect.
Fine application of digital twin technology in biomass boilers!
As biomass boilers are too complex to be efficiently handled by a human, a Control System is used. The installed system already increases a biomass boilers efficiency to a certain degree, but it can go higher. The Control System is not able to analyze fuel calorific value and water content of the fuel in the furnace, so it cannot effectively manage the burning process, but it does store the data needed to calculate what is happening in real-time. Control Systems do not cover and do not guaranty maximum efficiency in full range of Boiler operation and conditions.
Therefore, dedicated analytical systems based on Digital Twin and data analytic technology are used to simulate the burning process and find not only the optimal working conditions at any given time, but also gather data such as fuel calorific value and moisture content in fuel that is being burned right now...
https://blog.isa.org/how-digital-twin-technology-can-be-implemented-for-biomass-boilers
Dear Yuqian Lu,
my figure gives the answer to your question
Best regards
Anatol Badach
Dear Anatol Badach , please, bring us the link to the article. The only thing I can recognize is the application of Digital Twins in Internet of Things. Thanks.
Digital twin in process visualization!
The Digital Twin is a real-time visualization of all dynamic processes in your operations. Keep all movements on your shop floor in sight — via 2D or 3D visualization...
https://kinexon.com/products/kinexon-operating-system#whitepaper
Digital Twin and Simulation: Replicating Industrial Systems to Enable Improvement
Digital models, digital shadows, and digital twins are helping transform and optimize industrial and business operations...
Each type of digital twin involves a different relationship between the physical and digital worlds, and different data flows... (see figure).
https://www.isa.org/intech-home/2022/august-2022/features/digital-twin-and-simulation-replicating-industrial